From d57d1f69f4bca321d6365403f454a79a0bc20c26 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Wed, 12 Apr 2023 15:09:26 -0400
Subject: [PATCH 01/59] disk: add epoch handling to `u3_disk_init`

---
 pkg/vere/disk.c | 36 +++++++++++++++++++++++++++++++++---
 pkg/vere/vere.h |  3 ++-
 2 files changed, 35 insertions(+), 4 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 87c683fa9f..538238780e 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -867,7 +867,9 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
     c3_free(dir_c);
   }
 
-  //  create/load $pier/.urb/log, initialize db
+  //  create/load $pier/.urb/log
+  //  create/load $pier/.urb/log/0iN (epoch dir)
+  //  initialize db
   //
   {
     c3_c* log_c = c3_malloc(10 + strlen(pax_c));
@@ -882,6 +884,30 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
       return 0;
     }
 
+    //  iterate through the log directory to find the latest epoch,
+    //  or set it to "0i0" if none exist
+    //
+    u3_dent* den_u = log_u->com_u->all_u;
+    c3_c* lat_c;  //  latest epoch or "0i0" if none exist
+    if ( !den_u ) {
+      lat_c = "0i0";
+    } else {
+      while ( den_u ) {
+        if ( 0 == strncmp(den_u->nam_c, "0i", 2) ) {
+          //  update the lat_c if it's a later epoch
+          if ( !lat_c || (strcmp(den_u->nam_c, lat_c) > 0) ) {
+            lat_c = den_u->nam_c;
+          }
+        }
+        den_u = den_u->nex_u;
+      }
+    }
+
+    //  create/load $pier/.urb/log/0iN epoch directory
+    c3_c epo_c[8193];
+    snprintf(epo_c, sizeof(epo_c), "%s/%s", log_c, lat_c);
+    c3_mkdir(epo_c, 0700);
+
     //  Arbitrarily choosing 1TB as a "large enough" mapsize
     //
     //  per the LMDB docs:
@@ -895,12 +921,16 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
       #else
         0x10000000000;
       #endif
+      
+      fprintf(stderr, "disk: initializing database at %s\r\n", epo_c);
 
-      if ( 0 == (log_u->mdb_u = u3_lmdb_init(log_c, siz_i)) ) {
+      if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c, siz_i)) ) {
         fprintf(stderr, "disk: failed to initialize database\r\n");
-        c3_free(log_c);
         c3_free(log_u);
         return 0;
+      } else {
+        //  register the epoch directory in the disk struct
+        log_u->epo_u = u3_foil_folder(epo_c);
       }
     }
 
diff --git a/pkg/vere/vere.h b/pkg/vere/vere.h
index ca8a6da06d..ba06820a62 100644
--- a/pkg/vere/vere.h
+++ b/pkg/vere/vere.h
@@ -536,8 +536,9 @@
           u3_dire*         dir_u;               //  main pier directory
           u3_dire*         urb_u;               //  urbit system data
           u3_dire*         com_u;               //  log directory
+          u3_dire*         epo_u;               //  current epoch directory
           c3_o             liv_o;               //  live
-          void*            mdb_u;               //  lmdb environment.
+          void*            mdb_u;               //  lmdb environment
           c3_d             sen_d;               //  commit requested
           c3_d             dun_d;               //  committed
           u3_disk_cb        cb_u;               //  callbacks

From 4be7ec30102a409c7ac3a723e69653b97ce85009 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Thu, 13 Apr 2023 14:10:41 -0400
Subject: [PATCH 02/59] u3: add dir list to `u3_dire`

---
 pkg/vere/foil.c | 6 +++++-
 pkg/vere/vere.h | 3 ++-
 2 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/pkg/vere/foil.c b/pkg/vere/foil.c
index 26ff8b3d04..a241c32b66 100644
--- a/pkg/vere/foil.c
+++ b/pkg/vere/foil.c
@@ -106,13 +106,17 @@ u3_foil_folder(const c3_c* pax_c)
     dir_u = u3_dire_init(pax_c);
   }
 
-  /*  create entries for all files
+  /*  create entries for all files and directories
   */
   while ( UV_EOF != uv_fs_scandir_next(&ruq_u, &den_u) ) {
     if ( UV_DIRENT_FILE == den_u.type ) {
       u3_dent* det_u = u3_dent_init(den_u.name);
       det_u->nex_u = dir_u->all_u;
       dir_u->all_u = det_u;
+    } else if ( UV_DIRENT_DIR == den_u.type ) {
+      u3_dent* det_u = u3_dent_init(den_u.name);
+      det_u->nex_u = dir_u->dir_u;
+      dir_u->dir_u = det_u;
     }
   }
 
diff --git a/pkg/vere/vere.h b/pkg/vere/vere.h
index ba06820a62..a1522c6e40 100644
--- a/pkg/vere/vere.h
+++ b/pkg/vere/vere.h
@@ -130,6 +130,7 @@
         c3_c*    pax_c;                     //  path of directory
         uv_file  fil_u;                     //  file, opened read-only to fsync
         u3_dent* all_u;                     //  file list
+        u3_dent* dir_u;                     //  directory list
       } u3_dire;
 
     /* u3_save: checkpoint control.
@@ -538,7 +539,7 @@
           u3_dire*         com_u;               //  log directory
           u3_dire*         epo_u;               //  current epoch directory
           c3_o             liv_o;               //  live
-          void*            mdb_u;               //  lmdb environment
+          void*            mdb_u;               //  lmdb env of current epoch
           c3_d             sen_d;               //  commit requested
           c3_d             dun_d;               //  committed
           u3_disk_cb        cb_u;               //  callbacks

From 89e919865a7422d7e558c09600edeeb79d34961f Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Thu, 13 Apr 2023 15:02:20 -0400
Subject: [PATCH 03/59] disk: add `u3_disk_last_epoc` and `u3_disk_migrate`

---
 BUILD.bazel     |   4 +-
 pkg/vere/disk.c | 113 +++++++++++++++++++++++++++++++++++-------------
 pkg/vere/main.c |   4 +-
 pkg/vere/vere.h |  11 +++++
 4 files changed, 97 insertions(+), 35 deletions(-)

diff --git a/BUILD.bazel b/BUILD.bazel
index 4ea5807432..304ecb66d4 100644
--- a/BUILD.bazel
+++ b/BUILD.bazel
@@ -44,8 +44,8 @@ config_setting(
 # Version flag for clang.
 string_flag(
     name = "clang_version",
-    # macOS uses `clang-14.0.0` by default.
-    build_setting_default = "14.0.0",
+    # macOS uses `clang-14.0.3` by default.
+    build_setting_default = "14.0.3",
     visibility = ["//visibility:public"],
 )
 
diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 538238780e..867716ac74 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -2,6 +2,7 @@
 
 #include "noun.h"
 #include "vere.h"
+#include "version.h"
 #include "db/lmdb.h"
 
 struct _cd_read {
@@ -22,6 +23,12 @@ struct _cd_save {
   struct _u3_disk* log_u;
 };
 
+/* DISK FORMAT
+ */
+
+#define U3D_VER1   1
+#define U3D_VERLAT U3L_VER1
+
 #undef VERBOSE_DISK
 #undef DISK_TRACE_JAM
 #undef DISK_TRACE_CUE
@@ -868,8 +875,7 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
   }
 
   //  create/load $pier/.urb/log
-  //  create/load $pier/.urb/log/0iN (epoch dir)
-  //  initialize db
+  //  XX closures?
   //
   {
     c3_c* log_c = c3_malloc(10 + strlen(pax_c));
@@ -884,34 +890,26 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
       return 0;
     }
 
-    //  iterate through the log directory to find the latest epoch,
-    //  or set it to "0i0" if none exist
+    //  validation before migration
     //
-    u3_dent* den_u = log_u->com_u->all_u;
-    c3_c* lat_c;  //  latest epoch or "0i0" if none exist
-    if ( !den_u ) {
-      lat_c = "0i0";
-    } else {
-      while ( den_u ) {
-        if ( 0 == strncmp(den_u->nam_c, "0i", 2) ) {
-          //  update the lat_c if it's a later epoch
-          if ( !lat_c || (strcmp(den_u->nam_c, lat_c) > 0) ) {
-            lat_c = den_u->nam_c;
-          }
-        }
-        den_u = den_u->nex_u;
-      }
-    }
 
-    //  create/load $pier/.urb/log/0iN epoch directory
-    c3_c epo_c[8193];
-    snprintf(epo_c, sizeof(epo_c), "%s/%s", log_c, lat_c);
-    c3_mkdir(epo_c, 0700);
+    //  migrate to the correct disk format
+    //
+    u3_disk_migrate(log_u);
+
+    //  find the latest epoch directory
+    //
+    c3_d lat_d;
+    u3_disk_last_epoc(log_u, &lat_d);
+    fprintf(stderr, "disk: latest epoch is 0i%" PRIu64 "\r\n", lat_d);
 
-    //  Arbitrarily choosing 1TB as a "large enough" mapsize
+    //  load latest epoch directory and set it in log_u
     //
-    //  per the LMDB docs:
-    //  "[..] on 64-bit there is no penalty for making this huge (say 1TB)."
+    c3_c epo_c[8193];
+    snprintf(epo_c, 8192, "%s/0i%" PRIu64, log_c, lat_d);
+    log_u->epo_u = u3_foil_folder(epo_c);
+
+    //  initialize db of latest epoch
     //
     {
       const size_t siz_i =
@@ -922,15 +920,10 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
         0x10000000000;
       #endif
       
-      fprintf(stderr, "disk: initializing database at %s\r\n", epo_c);
-
       if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c, siz_i)) ) {
         fprintf(stderr, "disk: failed to initialize database\r\n");
         c3_free(log_u);
         return 0;
-      } else {
-        //  register the epoch directory in the disk struct
-        log_u->epo_u = u3_foil_folder(epo_c);
       }
     }
 
@@ -960,3 +953,61 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
 
   return log_u;
 }
+
+/* u3_disk_last_epoc: get latest epoch number.
+ */
+c3_d u3_disk_last_epoc(u3_disk* log_u, c3_d* lat_d) {
+  c3_d ret_d = 1;  //  return code 1 if no epoch directories exist
+  *lat_d = 0;      //  initialize lat_d to 0
+  u3_dent* den_u = log_u->com_u->dir_u;
+  while ( den_u ) {
+    c3_d epo_d = 0;
+    if ( 1 != sscanf(den_u->nam_c, "0i%" PRIu64, &epo_d) ) {
+      fprintf(stderr, "disk: epoch directory is not a @ui: %s\r\n", den_u->nam_c);
+    } else {
+      ret_d = 0;  //  return code 0 if at least one epoch directory exists
+    }
+    *lat_d = c3_max(epo_d, *lat_d);  //  update the latest epoch number
+    den_u = den_u->nex_u;
+  }
+
+  return ret_d;
+}
+
+/* u3_disk_migrate: migrates disk format.
+ */
+void u3_disk_migrate(u3_disk* log_u)
+{
+  c3_d lat_d;  //  latest epoch number
+
+  //  initialize disk v1 on fresh boots
+  //
+  if ( 1 == u3_disk_last_epoc(log_u, &lat_d) ) {  //  XX add another condition to ensure fresh boot
+    fprintf(stderr, "disk: migrating disk to v%d format\r\n", U3D_VER1);
+
+    //  create first epoch directory "0i0"
+    c3_c epo_c[8193];
+    snprintf(epo_c, sizeof(epo_c), "%s/0i0", log_u->com_u->pax_c);
+    c3_mkdir(epo_c, 0700);
+
+    //  create epoch version file
+    c3_c epv_c[8193];
+    snprintf(epv_c, sizeof(epv_c), "%s/epoc.txt", epo_c);
+    FILE* epv_f = fopen(epv_c, "a");
+    fprintf(epv_f, "%d\n", U3D_VER1);
+    fclose(epv_f);
+
+    //  create binary version file
+    c3_c biv_c[8193];
+    snprintf(biv_c, sizeof(biv_c), "%s/vere.txt", epo_c);
+    FILE* biv_f = fopen(biv_c, "a");
+    fprintf(biv_f, URBIT_VERSION);
+    fclose(biv_f);
+  }
+
+  //  migrate existing unversioned pier to v1
+  //
+  //  XX write this
+
+  return;
+}
\ No newline at end of file
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index 510ed95498..d2450b1cd9 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -2072,7 +2072,7 @@ _cw_chop(c3_i argc, c3_c* argv[])
   // gracefully shutdown the pier if it's running
   u3_disk* old_u = _cw_disk_init(u3_Host.dir_c);
 
-  // note: this include patch applications (if any)
+  // note: this includes patch applications (if any)
   u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
 
   // check if there's a *current* snapshot
@@ -2083,7 +2083,7 @@ _cw_chop(c3_i argc, c3_c* argv[])
     exit(1);
   }
 
-  if ( c3n == u3e_backup(c3y)) {  //  backup current snapshot
+  if ( c3n == u3e_backup(c3y) ) {  //  backup current snapshot
     fprintf(stderr, "chop: error: failed to backup snapshot\r\n");
     exit(1);
   }
diff --git a/pkg/vere/vere.h b/pkg/vere/vere.h
index a1522c6e40..a32d9a4604 100644
--- a/pkg/vere/vere.h
+++ b/pkg/vere/vere.h
@@ -8,6 +8,7 @@
 #include "noun.h"
 #include "serf.h"
 #include "uv.h"
+#include "version.h"
 
   /** Quasi-tunable parameters.
   **/
@@ -981,6 +982,16 @@
       */
         void
         u3_disk_plan(u3_disk* log_u, u3_fact* tac_u);
+      
+      /* u3_disk_last_epoc(): get latest epoch number.
+       */
+       c3_d
+       u3_disk_last_epoc(u3_disk* log_u, c3_d* lat_d);
+      
+      /* u3_disk_migrate(): migrates disk format.
+       */
+        void
+        u3_disk_migrate(u3_disk* log_u);
 
       /* u3_lord_init(): start serf.
       */

From 796de555b8f389961dba300d6dba5be027bc1fe3 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Fri, 14 Apr 2023 14:19:48 -0400
Subject: [PATCH 04/59] u3: make `u3e_backup` copy snapshot to arbitrary path

---
 pkg/noun/events.c | 38 ++++++++++++++++++++------------------
 pkg/noun/events.h |  4 ++--
 2 files changed, 22 insertions(+), 20 deletions(-)

diff --git a/pkg/noun/events.c b/pkg/noun/events.c
index 2d831ffef4..9d772faf9b 100644
--- a/pkg/noun/events.c
+++ b/pkg/noun/events.c
@@ -993,48 +993,48 @@ _ce_image_copy(u3e_image* fom_u, u3e_image* tou_u)
   return c3y;
 }
 
-/* u3e_backup();
+/* u3e_backup(): copy snapshot to pax_c, overwrite optionally.
 */
 c3_o
-u3e_backup(c3_o ovw_o)
+u3e_backup(c3_c* pax_c, c3_o ovw_o)
 {
   u3e_image nop_u = { .nam_c = "north", .pgs_w = 0 };
   u3e_image sop_u = { .nam_c = "south", .pgs_w = 0 };
   c3_i mod_i = O_RDWR | O_CREAT;
-  c3_c ful_c[8193];
 
-  snprintf(ful_c, 8192, "%s/.urb/bhk", u3P.dir_c);
+  if ( !pax_c ) {
+    fprintf(stderr, "loom: image backup: bad path\r\n");
+    return c3n;
+  }
 
-  if ( (c3n == ovw_o) && c3_mkdir(ful_c, 0700) ) {
+  if ( (c3n == ovw_o) && c3_mkdir(pax_c, 0700) ) {
     if ( EEXIST != errno ) {
       fprintf(stderr, "loom: image backup: %s\r\n", strerror(errno));
     }
     return c3n;
   }
 
-  snprintf(ful_c, 8192, "%s/.urb/bhk/%s.bin", u3P.dir_c, nop_u.nam_c);
+  c3_c nop_c[8193];
+  snprintf(nop_c, 8192, "%s/%s.bin", pax_c, nop_u.nam_c);
 
-  if ( -1 == (nop_u.fid_i = c3_open(ful_c, mod_i, 0666)) ) {
-    fprintf(stderr, "loom: c3_open %s: %s\r\n", ful_c, strerror(errno));
+  if ( -1 == (nop_u.fid_i = c3_open(nop_c, mod_i, 0666)) ) {
+    fprintf(stderr, "loom: c3_open %s: %s\r\n", nop_c, strerror(errno));
     return c3n;
   }
 
-  snprintf(ful_c, 8192, "%s/.urb/bhk/%s.bin", u3P.dir_c, sop_u.nam_c);
+  c3_c sop_c[8193];
+  snprintf(sop_c, 8192, "%s/%s.bin", pax_c, sop_u.nam_c);
 
-  if ( -1 == (sop_u.fid_i = c3_open(ful_c, mod_i, 0666)) ) {
-    fprintf(stderr, "loom: c3_open %s: %s\r\n", ful_c, strerror(errno));
+  if ( -1 == (sop_u.fid_i = c3_open(sop_c, mod_i, 0666)) ) {
+    fprintf(stderr, "loom: c3_open %s: %s\r\n", sop_c, strerror(errno));
     return c3n;
   }
 
   if (  (c3n == _ce_image_copy(&u3P.nor_u, &nop_u))
      || (c3n == _ce_image_copy(&u3P.sou_u, &sop_u)) )
   {
-
-    c3_unlink(ful_c);
-    snprintf(ful_c, 8192, "%s/.urb/bhk/%s.bin", u3P.dir_c, nop_u.nam_c);
-    c3_unlink(ful_c);
-    snprintf(ful_c, 8192, "%s/.urb/bhk", u3P.dir_c);
-    c3_rmdir(ful_c);
+    c3_unlink(nop_c);
+    c3_unlink(sop_c);
     fprintf(stderr, "loom: image backup failed\r\n");
     return c3n;
   }
@@ -1111,7 +1111,9 @@ u3e_save(void)
   _ce_patch_free(pat_u);
   _ce_patch_delete();
 
-  u3e_backup(c3n);
+  c3_c bhk_c[8193];
+  snprintf(bhk_c, sizeof(bhk_c), "%s/.urb/bhk", u3P.dir_c);
+  u3e_backup(bhk_c, c3n);
 }
 
 /* u3e_live(): start the checkpointing system.
diff --git a/pkg/noun/events.h b/pkg/noun/events.h
index b4c78fe364..a4d7c5cf43 100644
--- a/pkg/noun/events.h
+++ b/pkg/noun/events.h
@@ -64,10 +64,10 @@
 
   /** Functions.
   **/
-    /* u3e_backup(): copy the snapshot from chk to bhk. 
+    /* u3e_backup(): copy the snapshot to pax_c, overwrite optionally.
     */
       c3_o 
-      u3e_backup(c3_o ovw_o);
+      u3e_backup(c3_c* pax_c, c3_o ovw_o);
 
     /* u3e_fault(): handle a memory event with libsigsegv protocol.
     */

From 6d00b08ee3f6ffb94c1ca94ef2ad5870d7768c63 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Fri, 14 Apr 2023 14:22:33 -0400
Subject: [PATCH 05/59] disk: add `epoc` functions

---
 pkg/vere/disk.c | 140 +++++++++++++++++++++++++++++++++++-------------
 pkg/vere/vere.h |  10 ++--
 2 files changed, 110 insertions(+), 40 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 867716ac74..fee6ce28dd 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -893,24 +893,22 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
     //  validation before migration
     //
 
-    //  migrate to the correct disk format
+    //  migrate to the correct disk format if necessary
     //
     u3_disk_migrate(log_u);
 
-    //  find the latest epoch directory
+    //  initialize latest epoch
     //
     c3_d lat_d;
-    u3_disk_last_epoc(log_u, &lat_d);
+    u3_disk_epoc_last(log_u, &lat_d);
     fprintf(stderr, "disk: latest epoch is 0i%" PRIu64 "\r\n", lat_d);
 
-    //  load latest epoch directory and set it in log_u
+    //  initialize epoch's db
     //
     c3_c epo_c[8193];
     snprintf(epo_c, 8192, "%s/0i%" PRIu64, log_c, lat_d);
-    log_u->epo_u = u3_foil_folder(epo_c);
-
-    //  initialize db of latest epoch
-    //
+    fprintf(stderr, "epo_c: %s\r\n", epo_c);
+    c3_free(log_c);
     {
       const size_t siz_i =
       // 500 GiB is as large as musl on aarch64 wants to allow
@@ -919,15 +917,13 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
       #else
         0x10000000000;
       #endif
-      
+
       if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c, siz_i)) ) {
         fprintf(stderr, "disk: failed to initialize database\r\n");
         c3_free(log_u);
         return 0;
       }
     }
-
-    c3_free(log_c);
   }
 
   //  get the latest event number from the db
@@ -954,18 +950,101 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
   return log_u;
 }
 
-/* u3_disk_last_epoc: get latest epoch number.
- */
-c3_d u3_disk_last_epoc(u3_disk* log_u, c3_d* lat_d) {
-  c3_d ret_d = 1;  //  return code 1 if no epoch directories exist
-  *lat_d = 0;      //  initialize lat_d to 0
+/* u3_disk_epoc_init: create new epoch.
+*/
+c3_o u3_disk_epoc_init(u3_disk* log_u) {
+  //  set new epoch number
+  c3_d lat_d, new_d;
+  c3_o eps_o = u3_disk_epoc_last(log_u, &lat_d);
+  if ( c3n == eps_o ) {
+    new_d = 0;  //  no epochs yet, so create first one 0i0
+  } else {
+    //  ensure latest epoch isn't empty
+    c3_d fir_d, las_d;
+    if ( c3n == u3_lmdb_gulf(log_u->mdb_u, &fir_d, &las_d) ) {
+      fprintf(stderr, "disk: failed to get first/last event numbers\r\n");
+      return c3n;
+    }
+    if ( fir_d == las_d ) {
+      fprintf(stderr, "disk: latest epoch is empty; skipping epoch init\r\n");
+      return c3n;
+    }
+    new_d = 1 + lat_d;  //  create next epoch
+  }
+
+  //  create new epoch directory
+  c3_c epo_c[8193];
+  snprintf(epo_c, sizeof(epo_c), "%s/0i%" PRIu64, log_u->com_u->pax_c, new_d);
+  fprintf(stderr, "disk: creating new epoch directory 0i%" PRIu64 "\r\n", new_d);
+  if ( 0 != c3_mkdir(epo_c, 0700) ) {
+    fprintf(stderr, "disk: failed to create epoch directory\r\n");
+    return c3n;
+  }
+
+  //  create epoch version file
+  c3_c epv_c[8193];
+  snprintf(epv_c, sizeof(epv_c), "%s/epoc.txt", epo_c);
+  FILE* epv_f = fopen(epv_c, "a");
+  fprintf(epv_f, "%d\n", U3D_VER1);
+  fclose(epv_f);
+
+  //  create binary version file
+  c3_c biv_c[8193];
+  snprintf(biv_c, sizeof(biv_c), "%s/vere.txt", epo_c);
+  FILE* biv_f = fopen(biv_c, "a");
+  fprintf(biv_f, URBIT_VERSION);
+  fclose(biv_f);
+
+  //  copy snapshot (skip if first epoch 0i0)
+  if ( new_d > 0 ) {
+    if ( c3n == u3e_backup(epo_c, c3n) ) {
+      fprintf(stderr, "disk: failed to copy snapshot to new epoch\r\n");
+      goto fail;
+    }
+  }
+
+  //  initialize db of latest epoch
+  {
+    const size_t siz_i =
+    // 500 GiB is as large as musl on aarch64 wants to allow
+    #if (defined(U3_CPU_aarch64) && defined(U3_OS_linux))
+      0x7d00000000;
+    #else
+      0x10000000000;
+    #endif
+
+    if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c, siz_i)) ) {
+      fprintf(stderr, "disk: failed to initialize database\r\n");
+      c3_free(log_u);
+      goto fail;
+    }
+  }
+
+  //  load new epoch directory and set it in log_u
+  log_u->epo_u = u3_foil_folder(epo_c);
+
+  //  success
+  return c3y;
+
+fail:
+  c3_unlink(epv_c);
+  c3_unlink(biv_c);
+  c3_rmdir(epo_c);
+  return c3n;
+}
+
+/* u3_disk_epoc_last: get latest epoch number.
+*/
+c3_o u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d) {
+  c3_o ret_d = c3n;  //  return no if no epoch directories exist
+  *lat_d = 0;        //  initialize lat_d to 0
   u3_dent* den_u = log_u->com_u->dir_u;
   while ( den_u ) {
     c3_d epo_d = 0;
     if ( 1 != sscanf(den_u->nam_c, "0i%" PRIu64, &epo_d) ) {
       fprintf(stderr, "disk: epoch directory is not a @ui: %s\r\n", den_u->nam_c);
     } else {
-      ret_d = 0;  //  return code 0 if at least one epoch directory exists
+      ret_d = c3y;   //  return yes if at least one epoch directory exists
     }
     *lat_d = c3_max(epo_d, *lat_d);  //  update the latest epoch number
     den_u = den_u->nex_u;
@@ -982,30 +1061,17 @@ void u3_disk_migrate(u3_disk* log_u)
 
   //  initialize disk v1 on fresh boots
   //
-  if ( 1 == u3_disk_last_epoc(log_u, &lat_d) ) {  //  XX add another condition to ensure fresh boot
+  if ( c3n == u3_disk_epoc_last(log_u, &lat_d) ) {  //  XX ensure fresh boot
     fprintf(stderr, "disk: migrating disk to v%d format\r\n", U3D_VER1);
 
-    //  create first epoch directory "0i0"
-    c3_c epo_c[8193];
-    snprintf(epo_c, sizeof(epo_c), "%s/0i0", log_u->com_u->pax_c);
-    c3_mkdir(epo_c, 0700);
-
-    //  create epoch version file
-    c3_c epv_c[8193];
-    snprintf(epv_c, sizeof(epv_c), "%s/epoc.txt", epo_c);
-    FILE* epv_f = fopen(epv_c, "a");
-    fprintf(epv_f, "%d\n", U3D_VER1);
-    fclose(epv_f);
-
-    //  create binary version file
-    c3_c biv_c[8193];
-    snprintf(biv_c, sizeof(biv_c), "%s/vere.txt", epo_c);
-    FILE* biv_f = fopen(biv_c, "a");
-    fprintf(biv_f, URBIT_VERSION);
-    fclose(biv_f);
+    //  initialize first epoch "0i0"
+    if ( c3n == u3_disk_epoc_init(log_u) ) {
+      fprintf(stderr, "disk: migrate: failed to initialize first epoch\r\n");
+      exit(1);
+    }
   }
 
-  //  migrate existing unversioned pier to v1
+  //  migrate existing, unversioned disk to v1
   //
   //  XX write this
 
diff --git a/pkg/vere/vere.h b/pkg/vere/vere.h
index a32d9a4604..694769cf57 100644
--- a/pkg/vere/vere.h
+++ b/pkg/vere/vere.h
@@ -8,7 +8,6 @@
 #include "noun.h"
 #include "serf.h"
 #include "uv.h"
-#include "version.h"
 
   /** Quasi-tunable parameters.
   **/
@@ -983,10 +982,15 @@
         void
         u3_disk_plan(u3_disk* log_u, u3_fact* tac_u);
       
+      /* u3_disk_init_epoc(): create new epoch.
+       */
+       c3_o
+       u3_disk_epoc_init(u3_disk* log_u);
+      
       /* u3_disk_last_epoc(): get latest epoch number.
        */
-       c3_d
-       u3_disk_last_epoc(u3_disk* log_u, c3_d* lat_d);
+       c3_o
+       u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d);
       
       /* u3_disk_migrate(): migrates disk format.
        */

From 488adc3f7cede004a09b18796386ae03409b6d3c Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Mon, 17 Apr 2023 05:13:11 -0400
Subject: [PATCH 06/59] cli: add `roll` subcommand for epoch rollovers

---
 pkg/noun/events.h |  2 +-
 pkg/vere/disk.c   | 72 +++++++++++++++++++++++++++-------------
 pkg/vere/foil.c   |  4 +--
 pkg/vere/main.c   | 84 +++++++++++++++++++++++++++++++++++++++++++++--
 pkg/vere/vere.h   |  2 +-
 pkg/vere/ward.c   | 12 +++++++
 6 files changed, 147 insertions(+), 29 deletions(-)

diff --git a/pkg/noun/events.h b/pkg/noun/events.h
index a4d7c5cf43..058fa2fe23 100644
--- a/pkg/noun/events.h
+++ b/pkg/noun/events.h
@@ -64,7 +64,7 @@
 
   /** Functions.
   **/
-    /* u3e_backup(): copy the snapshot to pax_c, overwrite optionally.
+    /* u3e_backup(): copy the snapshot to pax_c, overwrite optional.
     */
       c3_o 
       u3e_backup(c3_c* pax_c, c3_o ovw_o);
diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index fee6ce28dd..8eefc7c973 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -890,26 +890,25 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
       return 0;
     }
 
-    //  validation before migration
+    //  XX validation before migration
     //
 
     //  migrate to the correct disk format if necessary
     //
     u3_disk_migrate(log_u);
 
-    //  initialize latest epoch
+    //  get latest epoch number
     //
     c3_d lat_d;
     u3_disk_epoc_last(log_u, &lat_d);
-    fprintf(stderr, "disk: latest epoch is 0i%" PRIu64 "\r\n", lat_d);
 
     //  initialize epoch's db
     //
     c3_c epo_c[8193];
     snprintf(epo_c, 8192, "%s/0i%" PRIu64, log_c, lat_d);
-    fprintf(stderr, "epo_c: %s\r\n", epo_c);
     c3_free(log_c);
     {
+      // XX extract into function?
       const size_t siz_i =
       // 500 GiB is as large as musl on aarch64 wants to allow
       #if (defined(U3_CPU_aarch64) && defined(U3_OS_linux))
@@ -924,24 +923,33 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
         return 0;
       }
     }
-  }
+    fprintf(stderr, "disk: loaded epoch 0i%" PRIu64 "\r\n", lat_d);
 
-  //  get the latest event number from the db
-  //
-  {
-    log_u->dun_d = 0;
-    c3_d fir_d;
-
-    if ( c3n == u3_lmdb_gulf(log_u->mdb_u, &fir_d, &log_u->dun_d) ) {
-      fprintf(stderr, "disk: failed to load latest event from database\r\n");
-      c3_free(log_u);
+    //  get first/last event numbers from lmdb
+    c3_d fir_d, las_d;
+    if ( c3n == u3_lmdb_gulf(log_u->mdb_u, &fir_d, &las_d) ) {
+      fprintf(stderr, "disk: failed to get first/last event numbers\r\n");
       return 0;
     }
 
-    log_u->sen_d = log_u->dun_d;
+    //  initialize dun_d/sen_d values
+    if ( 0 == las_d ) {               //  fresh epoch (no events in lmdb yet)
+      if ( 0 == lat_d ) {             //  first epoch
+        log_u->dun_d = 0;
+        log_u->sen_d = 0;
+      } else {                        //  not first epoch
+        log_u->dun_d = lat_d - 1;     //  set dun_d to last event in prev epoch
+        log_u->sen_d = log_u->dun_d;
+      }
+    } else {                          //  not fresh epoch            
+      log_u->dun_d = las_d;           //  set dun_d to last event in lmdb        
+      log_u->sen_d = las_d;
+    }
+
+    //  mark the log as live
+    log_u->liv_o = c3y;
   }
 
-  log_u->liv_o = c3y;
 
 #if defined(DISK_TRACE_JAM) || defined(DISK_TRACE_CUE)
   u3t_trace_open(pax_c);
@@ -966,16 +974,15 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
       return c3n;
     }
     if ( fir_d == las_d ) {
-      fprintf(stderr, "disk: latest epoch is empty; skipping epoch init\r\n");
+      fprintf(stderr, "disk: latest epoch is empty; skipping rollover\r\n");
       return c3n;
     }
-    new_d = 1 + lat_d;  //  create next epoch
+    new_d = 1 + las_d;  //  create next epoch
   }
 
   //  create new epoch directory
   c3_c epo_c[8193];
   snprintf(epo_c, sizeof(epo_c), "%s/0i%" PRIu64, log_u->com_u->pax_c, new_d);
-  fprintf(stderr, "disk: creating new epoch directory 0i%" PRIu64 "\r\n", new_d);
   if ( 0 != c3_mkdir(epo_c, 0700) ) {
     fprintf(stderr, "disk: failed to create epoch directory\r\n");
     return c3n;
@@ -995,15 +1002,26 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
   fprintf(biv_f, URBIT_VERSION);
   fclose(biv_f);
 
-  //  copy snapshot (skip if first epoch 0i0)
+  //  copy snapshot (skip if first epoch)
   if ( new_d > 0 ) {
-    if ( c3n == u3e_backup(epo_c, c3n) ) {
+    if ( c3n == u3e_backup(epo_c, c3y) ) {
       fprintf(stderr, "disk: failed to copy snapshot to new epoch\r\n");
       goto fail;
     }
   }
 
-  //  initialize db of latest epoch
+  //  get metadata from old epoch's db
+  c3_d     who_d[2];
+  c3_o     fak_o;
+  c3_w     lif_w;
+  if ( c3y == eps_o ) {
+    if ( c3y != u3_disk_read_meta(log_u->mdb_u, who_d, &fak_o, &lif_w) ) {
+      fprintf(stderr, "disk: failed to read metadata\r\n");
+      goto fail;
+    }
+  }
+
+  //  initialize db of new epoch
   {
     const size_t siz_i =
     // 500 GiB is as large as musl on aarch64 wants to allow
@@ -1020,6 +1038,14 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
     }
   }
 
+  // write the metadata to the database
+  if ( c3y == eps_o ) {
+    if ( c3n == u3_disk_save_meta(log_u->mdb_u, who_d, fak_o, lif_w) ) {
+      fprintf(stderr, "disk: failed to save metadata\r\n");
+      exit(1);
+    }
+  }
+
   //  load new epoch directory and set it in log_u
   log_u->epo_u = u3_foil_folder(epo_c);
 
@@ -1038,7 +1064,7 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
 c3_o u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d) {
   c3_o ret_d = c3n;  //  return no if no epoch directories exist
   *lat_d = 0;        //  initialize lat_d to 0
-  u3_dent* den_u = log_u->com_u->dir_u;
+  u3_dent* den_u = log_u->com_u->dil_u;
   while ( den_u ) {
     c3_d epo_d = 0;
     if ( 1 != sscanf(den_u->nam_c, "0i%" PRIu64, &epo_d) ) {
diff --git a/pkg/vere/foil.c b/pkg/vere/foil.c
index a241c32b66..9fae86973f 100644
--- a/pkg/vere/foil.c
+++ b/pkg/vere/foil.c
@@ -115,8 +115,8 @@ u3_foil_folder(const c3_c* pax_c)
       dir_u->all_u = det_u;
     } else if ( UV_DIRENT_DIR == den_u.type ) {
       u3_dent* det_u = u3_dent_init(den_u.name);
-      det_u->nex_u = dir_u->dir_u;
-      dir_u->dir_u = det_u;
+      det_u->nex_u = dir_u->dil_u;
+      dir_u->dil_u = det_u;
     }
   }
 
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index d2450b1cd9..6aba090477 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -682,6 +682,7 @@ _cw_usage(c3_c* bin_c)
     "  %s next %.*s              request upgrade:\n",
     "  %s queu %.*s<at-event>    cue state:\n",
     "  %s chop %.*s              truncate event log:\n",
+    "  %s roll %.*s              rollover to new epoch:\n",
     "  %s vere ARGS <output dir>    download binary:\n",
     "\n  run as a 'serf':\n",
     "    %s serf <pier> <key> <flags> <cache-size> <at-event>"
@@ -1151,7 +1152,7 @@ _cw_disk_init(c3_c* dir_c)
   u3_disk*  log_u = u3_disk_init(dir_c, cb_u);
 
   if ( !log_u ) {
-    fprintf(stderr, "unable to open event log\n");
+    fprintf(stderr, "unable to open event log\n");;
     exit(1);
   }
 
@@ -2083,7 +2084,9 @@ _cw_chop(c3_i argc, c3_c* argv[])
     exit(1);
   }
 
-  if ( c3n == u3e_backup(c3y) ) {  //  backup current snapshot
+  c3_c bhk_c[8193];
+  snprintf(bhk_c, sizeof(bhk_c), "%s/.urb/bhk", u3_Host.dir_c);
+  if ( c3n == u3e_backup(bhk_c, c3y) ) {  //  backup current snapshot
     fprintf(stderr, "chop: error: failed to backup snapshot\r\n");
     exit(1);
   }
@@ -2189,6 +2192,82 @@ _cw_chop(c3_i argc, c3_c* argv[])
   fprintf(stderr, "      `mv %s %s` then try again\r\n", bak_c, dat_c);
 }
 
+/* _cw_roll(): rollover to new epoch
+ */
+static void
+_cw_roll(c3_i argc, c3_c* argv[])
+{
+  c3_i ch_i, lid_i;
+  c3_w arg_w;
+
+  static struct option lop_u[] = {
+    { "loom", required_argument, NULL, c3__loom },
+    { NULL, 0, NULL, 0 }
+  };
+
+  u3_Host.dir_c = _main_pier_run(argv[0]);
+
+  while ( -1 != (ch_i=getopt_long(argc, argv, "", lop_u, &lid_i)) ) {
+    switch ( ch_i ) {
+      case c3__loom: {
+        if (_main_readw_loom("loom", &u3_Host.ops_u.lom_y)) {
+          exit(1);
+        }
+      } break;
+
+      case '?': {
+        fprintf(stderr, "invalid argument\r\n");
+        exit(1);
+      } break;
+    }
+  }
+
+  //  argv[optind] is always "roll"
+  //
+
+  if ( !u3_Host.dir_c ) {
+    if ( optind + 1 < argc ) {
+      u3_Host.dir_c = argv[optind + 1];
+    }
+    else {
+      fprintf(stderr, "invalid command, pier required\r\n");
+      exit(1);
+    }
+
+    optind++;
+  }
+
+  if ( optind + 1 != argc ) {
+    fprintf(stderr, "invalid command\r\n");
+    exit(1);
+  }
+
+  // gracefully shutdown the pier if it's running
+  u3_disk* log_u = _cw_disk_init(u3_Host.dir_c);
+
+  // note: this includes patch applications (if any)
+  u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
+
+  // check if there's a *current* snapshot
+  if ( log_u->dun_d != u3A->eve_d ) {
+    fprintf(stderr, "roll: error: snapshot is out of date, please "
+                    "start/shutdown your pier gracefully first\r\n");
+    fprintf(stderr, "roll: eve_d: %" PRIu64 ", dun_d: %" PRIu64 "\r\n", \
+                     u3A->eve_d, log_u->dun_d);
+    exit(1);
+  }
+
+  //  create new epoch
+  if ( c3n == u3_disk_epoc_init(log_u) ) {
+    fprintf(stderr, "roll: error: failed to initialize new epoch\r\n");
+    exit(1);
+  }
+
+  //  report success
+  c3_d epo_d = log_u->dun_d + 1;
+  fprintf(stderr, "roll: success: created epoch 0i%" PRIu64 "\r\n", epo_d);
+}
+
 /* _cw_vere(): download vere
 */
 static void
@@ -2451,6 +2530,7 @@ _cw_utils(c3_i argc, c3_c* argv[])
     case c3__prep: _cw_prep(argc, argv); return 2; // continue on
     case c3__queu: _cw_queu(argc, argv); return 1;
     case c3__chop: _cw_chop(argc, argv); return 1;
+    case c3__roll: _cw_roll(argc, argv); return 1;
     case c3__vere: _cw_vere(argc, argv); return 1;
     case c3__vile: _cw_vile(argc, argv); return 1;
 
diff --git a/pkg/vere/vere.h b/pkg/vere/vere.h
index 694769cf57..fde11bb193 100644
--- a/pkg/vere/vere.h
+++ b/pkg/vere/vere.h
@@ -130,7 +130,7 @@
         c3_c*    pax_c;                     //  path of directory
         uv_file  fil_u;                     //  file, opened read-only to fsync
         u3_dent* all_u;                     //  file list
-        u3_dent* dir_u;                     //  directory list
+        u3_dent* dil_u;                     //  directory list
       } u3_dire;
 
     /* u3_save: checkpoint control.
diff --git a/pkg/vere/ward.c b/pkg/vere/ward.c
index 4c90be7179..9d6c8cbed3 100644
--- a/pkg/vere/ward.c
+++ b/pkg/vere/ward.c
@@ -39,6 +39,7 @@ u3_dire_init(const c3_c* pax_c)
 {
   u3_dire *dir_u = c3_malloc(sizeof *dir_u);
   dir_u->all_u = 0;
+  dir_u->dil_u = 0;
   dir_u->pax_c = c3_malloc(1 + strlen(pax_c));
   strcpy(dir_u->pax_c, pax_c);
 
@@ -61,6 +62,17 @@ u3_dire_free(u3_dire *dir_u)
     }
   }
 
+  {
+    u3_dent *det_u = dir_u->dil_u;
+    u3_dent *nex_u;
+
+    while ( det_u ) {
+      nex_u = det_u->nex_u;
+      u3_dent_free(det_u);
+      det_u = nex_u;
+    }
+  }
+
   c3_free(dir_u->pax_c);
   c3_free(dir_u);
 }

From f0289f989d350692246d29a00a1694de68677659 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Mon, 17 Apr 2023 12:45:38 -0400
Subject: [PATCH 07/59] cli: make `chop` work with epochs

---
 pkg/vere/disk.c |  30 ++++++++++++
 pkg/vere/main.c | 118 ++++++++++--------------------------------------
 pkg/vere/vere.h |   9 +++-
 3 files changed, 60 insertions(+), 97 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 8eefc7c973..9aa0f68707 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -1059,6 +1059,36 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
   return c3n;
 }
 
+/* u3_disk_epoc_kill: delete an epoch.
+*/
+c3_o u3_disk_epoc_kill(u3_disk* log_u, c3_d epo_d) {
+  //  get epoch directory
+  c3_c epo_c[8193];
+  snprintf(epo_c, sizeof(epo_c), "%s/0i%" PRIu64, log_u->com_u->pax_c, epo_d);
+
+  //  delete files in epoch directory
+  u3_dire* dir_u = u3_foil_folder(epo_c);
+  u3_dent* den_u = dir_u->all_u;
+  while ( den_u ) {
+    c3_c fil_c[8193];
+    snprintf(fil_c, sizeof(fil_c), "%s/%s", epo_c, den_u->nam_c);
+    if ( 0 != c3_unlink(fil_c) ) {
+      fprintf(stderr, "disk: failed to delete file in epoch directory\r\n");
+      return c3n;
+    }
+    den_u = den_u->nex_u;
+  }
+
+  //  delete epoch directory
+  if ( 0 != c3_rmdir(epo_c) ) {
+    fprintf(stderr, "disk: failed to delete epoch directory\r\n");
+    return c3n;
+  }
+
+  //  success
+  return c3y;
+}
+
 /* u3_disk_epoc_last: get latest epoch number.
 */
 c3_o u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d) {
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index 6aba090477..334a9dddd6 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -2071,125 +2071,53 @@ _cw_chop(c3_i argc, c3_c* argv[])
   }
 
   // gracefully shutdown the pier if it's running
-  u3_disk* old_u = _cw_disk_init(u3_Host.dir_c);
+  u3_disk* log_u = _cw_disk_init(u3_Host.dir_c);
 
   // note: this includes patch applications (if any)
   u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
 
   // check if there's a *current* snapshot
-  if ( old_u->dun_d != u3A->eve_d ) {
+  if ( log_u->dun_d != u3A->eve_d ) {
     fprintf(stderr, "chop: error: snapshot is out of date, please "
                     "start/shutdown your pier gracefully first\r\n");
-    fprintf(stderr, "chop: eve_d: %" PRIu64 ", dun_d: %" PRIu64 "\r\n", u3A->eve_d, old_u->dun_d);
-    exit(1);
-  }
-
-  c3_c bhk_c[8193];
-  snprintf(bhk_c, sizeof(bhk_c), "%s/.urb/bhk", u3_Host.dir_c);
-  if ( c3n == u3e_backup(bhk_c, c3y) ) {  //  backup current snapshot
-    fprintf(stderr, "chop: error: failed to backup snapshot\r\n");
-    exit(1);
-  }
-
-  // initialize the lmdb environment
-  // see disk.c:885
-  const size_t siz_i =
-  // 500 GiB is as large as musl on aarch64 wants to allow
-  #if (defined(U3_CPU_aarch64) && defined(U3_OS_linux))
-    0x7d00000000;
-  #else
-    0x10000000000;
-  #endif
-  c3_c log_c[8193];
-  snprintf(log_c, sizeof(log_c), "%s/.urb/log", u3_Host.dir_c);
-
-  // get the first/last event numbers from the event log
-  c3_d fir_d, las_d;
-  if ( c3n == u3_lmdb_gulf(old_u->mdb_u, &fir_d, &las_d) ) {
-    fprintf(stderr, "chop: failed to load latest event from database\r\n");
+    fprintf(stderr, "chop: eve_d: %" PRIu64 ", dun_d: %" PRIu64 "\r\n", u3A->eve_d, log_u->dun_d);
     exit(1);
   }
 
-  // get the metadata
-  c3_d     who_d[2];
-  c3_o     fak_o;
-  c3_w     lif_w;
-  if ( c3y != u3_disk_read_meta(old_u->mdb_u, who_d, &fak_o, &lif_w) ) {
-    fprintf(stderr, "chop: failed to read metadata\r\n");
+  //  create new epoch
+  if ( c3n == u3_disk_epoc_init(log_u) ) {
+    fprintf(stderr, "chop: failed to create new epoch\r\n");
     exit(1);
   }
 
-  // get the last event
-  u3_lmdb_walk itr_u;
-  size_t       len_i;
-  void*        buf_v[1];
-  if ( c3n == u3_lmdb_walk_init(old_u->mdb_u, &itr_u, las_d, las_d) ) {
-    fprintf(stderr, "chop: failed to initialize iterator\r\n");
-    exit(1);
-  }
-  if ( c3n == u3_lmdb_walk_next(&itr_u, &len_i, buf_v) ) {
-    fprintf(stderr, "chop: failed to read event\r\n");
+  //  get latest epoch number
+  c3_d lat_d;
+  if ( c3n == u3_disk_epoc_last(log_u, &lat_d) ) {
+    fprintf(stderr, "chop: failed to find last epoch\r\n");
     exit(1);
   }
-  u3_lmdb_walk_done(&itr_u);
 
-  // initialize a fresh lmdb environment in the "chop" subdir
-  c3_c cho_c[8193];
-  snprintf(cho_c, sizeof(cho_c), "%s/chop", log_c);
-  if ( 0 != access(cho_c, F_OK) ) {
-    if ( 0 != c3_mkdir(cho_c, 0700) ) {
-      fprintf(stderr, "chop: failed to create chop directory\r\n");
-      exit(1);
+  //  delete all but the last epoch
+  u3_dent* den_u = log_u->com_u->dil_u;
+  c3_d epo_d = 0;
+  while ( den_u && epo_d <= lat_d ) {
+    if ( 1 != sscanf(den_u->nam_c, "0i%" PRIu64, &epo_d) ) {
+      fprintf(stderr, "disk: epoch directory is not a @ui: %s\r\n", den_u->nam_c);
+    } else {
+      fprintf(stderr, "chop: deleting epoch 0i%" PRIu64 "\r\n", epo_d);
+      if ( c3y != u3_disk_epoc_kill(log_u, epo_d) ) {
+        fprintf(stderr, "chop: failed to delete epoch 0i%" PRIu64 "\r\n", epo_d);
+      }
     }
-  }
-  MDB_env* new_u = u3_lmdb_init(cho_c, siz_i);
-  if ( !new_u ) {
-    fprintf(stderr, "chop: failed to initialize new database\r\n");
-    exit(1);
-  }
-
-  // write the metadata to the database
-  if ( c3n == u3_disk_save_meta(new_u, who_d, fak_o, lif_w) ) {
-    fprintf(stderr, "chop: failed to save metadata\r\n");
-    exit(1);
-  }
-
-  // write the last event to the database
-  // warning: this relies on the old database still being open
-  if ( c3n == u3_lmdb_save(new_u, las_d, 1, buf_v, &len_i) ) {
-    fprintf(stderr, "chop: failed to write last event\r\n");
-    exit(1);
-  }
-
-  // backup the original database file
-  c3_c dat_c[8193], bak_c[8193];
-  snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", log_c);
-  // "data_<first>-<last>.mdb.bak"
-  snprintf(bak_c, sizeof(bak_c), "%s/data_%" PRIu64 "-%" PRIu64 ".mdb.bak", cho_c, fir_d, las_d);
-  if ( 0 != c3_rename(dat_c, bak_c) ) {
-    fprintf(stderr, "chop: failed to backup original database file\r\n");
-    exit(1);
-  }
-
-  // rename new database file to be official
-  c3_c new_c[8193];
-  snprintf(new_c, sizeof(new_c), "%s/data.mdb", cho_c);
-  if ( 0 != c3_rename(new_c, dat_c) ) {
-    fprintf(stderr, "chop: failed to rename new database file\r\n");
-    exit(1);
+    den_u = den_u->nex_u;
   }
 
   // cleanup
-  u3_disk_exit(old_u);
-  u3_lmdb_exit(new_u);
+  u3_disk_exit(log_u);
   u3m_stop();
 
   // success
   fprintf(stderr, "chop: event log truncation complete\r\n");
-  fprintf(stderr, "      event log backup written to %s\r\n", bak_c);
-  fprintf(stderr, "      WARNING: ENSURE YOU CAN RESTART YOUR SHIP BEFORE DELETING YOUR EVENT LOG BACKUP FILE!\r\n");
-  fprintf(stderr, "      if you can't, restore your log by running:\r\n");
-  fprintf(stderr, "      `mv %s %s` then try again\r\n", bak_c, dat_c);
 }
 
 /* _cw_roll(): rollover to new epoch
diff --git a/pkg/vere/vere.h b/pkg/vere/vere.h
index fde11bb193..a2726d2341 100644
--- a/pkg/vere/vere.h
+++ b/pkg/vere/vere.h
@@ -982,12 +982,17 @@
         void
         u3_disk_plan(u3_disk* log_u, u3_fact* tac_u);
       
-      /* u3_disk_init_epoc(): create new epoch.
+      /* u3_disk_epoc_init(): create new epoch.
        */
        c3_o
        u3_disk_epoc_init(u3_disk* log_u);
       
-      /* u3_disk_last_epoc(): get latest epoch number.
+      /* u3_disk_epoc_kill(): delete an epoch.
+       */
+       c3_o
+       u3_disk_epoc_kill(u3_disk* log_u, c3_d epo_d);
+      
+      /* u3_disk_epoc_last(): get latest epoch number.
        */
        c3_o
        u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d);

From c54d6c0bba8dfa100705322a40536dd4695e1499 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Tue, 18 Apr 2023 15:23:03 -0400
Subject: [PATCH 08/59] disk: migrate existing piers to v1 format (epochs)

---
 pkg/vere/disk.c | 123 ++++++++++++++++++++++++++++++++++++++++++------
 pkg/vere/main.c |   6 +--
 2 files changed, 112 insertions(+), 17 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 9aa0f68707..72dd3fcf91 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -891,19 +891,15 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
     }
 
     //  XX validation before migration
-    //
 
     //  migrate to the correct disk format if necessary
-    //
     u3_disk_migrate(log_u);
 
     //  get latest epoch number
-    //
     c3_d lat_d;
     u3_disk_epoc_last(log_u, &lat_d);
 
     //  initialize epoch's db
-    //
     c3_c epo_c[8193];
     snprintf(epo_c, 8192, "%s/0i%" PRIu64, log_c, lat_d);
     c3_free(log_c);
@@ -984,7 +980,7 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
   c3_c epo_c[8193];
   snprintf(epo_c, sizeof(epo_c), "%s/0i%" PRIu64, log_u->com_u->pax_c, new_d);
   if ( 0 != c3_mkdir(epo_c, 0700) ) {
-    fprintf(stderr, "disk: failed to create epoch directory\r\n");
+    fprintf(stderr, "disk: failed to create epoch directory %" PRIu64 "\r\n", new_d);
     return c3n;
   }
 
@@ -1094,7 +1090,7 @@ c3_o u3_disk_epoc_kill(u3_disk* log_u, c3_d epo_d) {
 c3_o u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d) {
   c3_o ret_d = c3n;  //  return no if no epoch directories exist
   *lat_d = 0;        //  initialize lat_d to 0
-  u3_dent* den_u = log_u->com_u->dil_u;
+  u3_dent* den_u = u3_foil_folder(log_u->com_u->pax_c)->dil_u;
   while ( den_u ) {
     c3_d epo_d = 0;
     if ( 1 != sscanf(den_u->nam_c, "0i%" PRIu64, &epo_d) ) {
@@ -1113,23 +1109,122 @@ c3_o u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d) {
  */
 void u3_disk_migrate(u3_disk* log_u)
 {
-  c3_d lat_d;  //  latest epoch number
+  //  check if data.mdb exists in log directory
+  c3_o dat_o = c3n;
+  c3_c dat_c[8193];
+  snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", log_u->com_u->pax_c);
+  if ( 0 == access(dat_c, F_OK) ) {
+    dat_o = c3y;
+  }
 
-  //  initialize disk v1 on fresh boots
-  //
-  if ( c3n == u3_disk_epoc_last(log_u, &lat_d) ) {  //  XX ensure fresh boot
+  //  check if any epochs exist
+  c3_d lat_d;
+  c3_o epo_o = u3_disk_epoc_last(log_u, &lat_d);
+
+  if ( c3n == epo_o && c3n == dat_o ) {
+    //  initialize disk v1 on fresh boots
+    fprintf(stderr, "disk: initializing disk with v%d format\r\n", U3D_VER1);
+
+    //  initialize first epoch "0i0"
+    if ( c3n == u3_disk_epoc_init(log_u) ) {
+      fprintf(stderr, "disk: migrate: failed to initialize first epoch\r\n");
+      exit(1);
+    }
+  } else if ( c3n == epo_o ) {
+    //  migrate existing, unversioned disk to v1
     fprintf(stderr, "disk: migrating disk to v%d format\r\n", U3D_VER1);
 
+    //  initialize pre-migrated lmdb
+    MDB_env* old_u;
+    {
+      const size_t siz_i =
+      // 500 GiB is as large as musl on aarch64 wants to allow
+      #if (defined(U3_CPU_aarch64) && defined(U3_OS_linux))
+        0x7d00000000;
+      #else
+        0x10000000000;
+      #endif
+
+      if ( 0 == (old_u = u3_lmdb_init(log_u->com_u->pax_c, siz_i)) ) {
+        fprintf(stderr, "disk: failed to initialize database\r\n");
+        c3_free(log_u);
+        return;
+      }
+    }
+
+    //  get first/last event numbers from pre-migrated lmdb
+    c3_d fir_d, las_d;
+    if ( c3n == u3_lmdb_gulf(old_u, &fir_d, &las_d) ) {
+      fprintf(stderr, "disk: failed to get first/last event numbers\r\n");
+      return;
+    }
+
+    //  boot
+    u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
+
+    //  make sure there's a current snapshot
+    if ( las_d != u3A->eve_d ) {
+      fprintf(stderr, "disk: migrate: error: snapshot is out of date, please "
+                      "start/shutdown your pier gracefully first\r\n");
+      exit(1);
+    }
+
+    //  shutdown pre-migrated u3 system and lmdb
+    u3m_stop();
+
     //  initialize first epoch "0i0"
     if ( c3n == u3_disk_epoc_init(log_u) ) {
       fprintf(stderr, "disk: migrate: failed to initialize first epoch\r\n");
       exit(1);
     }
-  }
 
-  //  migrate existing, unversioned disk to v1
-  //
-  //  XX write this
+    //  move data.mdb to 0i0/data.mdb
+    c3_c new_c[8193];
+    snprintf(new_c, sizeof(new_c), "%s/0i0/data.mdb", log_u->com_u->pax_c);
+    if ( 0 != rename(dat_c, new_c) ) {
+      fprintf(stderr, "disk: migrate: failed to move data.mdb\r\n");
+      exit(1);
+    }
+
+    //  move lock.mdb to 0i0/lock.mdb
+    c3_c old_c[8193];
+    snprintf(old_c, sizeof(old_c), "%s/lock.mdb", log_u->com_u->pax_c);
+    snprintf(new_c, sizeof(new_c), "%s/0i0/lock.mdb", log_u->com_u->pax_c);
+    if ( 0 != rename(old_c, new_c) ) {
+      fprintf(stderr, "disk: migrate: failed to move lock.mdb\r\n");
+      exit(1);
+    }
+
+    //  rollover to new epoch
+    log_u->mdb_u = old_u;
+    if ( c3n == u3_disk_epoc_init(log_u) ) {
+      fprintf(stderr, "roll: error: failed to initialize new epoch\r\n");
+      exit(1);
+    }
+
+    //  delete backup snapshot
+    c3_c bhk_c[8193];
+    snprintf(bhk_c, sizeof(bhk_c), "%s/.urb/bhk", u3_Host.dir_c);
+    c3_c nop_c[8193];
+    snprintf(nop_c, sizeof(nop_c), "%s/north.bin", bhk_c);
+    c3_c sop_c[8193];
+    snprintf(sop_c, sizeof(sop_c), "%s/south.bin", bhk_c);
+    if ( c3n == c3_unlink(nop_c) ) {
+      fprintf(stderr, "disk: migrate: failed to delete north.bin\r\n");
+      exit(1);
+    } else if ( c3n == c3_unlink(sop_c) ) {
+      fprintf(stderr, "disk: migrate: failed to delete south.bin\r\n");
+      exit(1);
+    } else {
+      if ( c3n == c3_rmdir(bhk_c) ) {
+        fprintf(stderr, "disk: migrate: failed to delete bhk\r\n");
+        exit(1);
+      }
+    }
+
+    //  success
+    fprintf(stderr, "disk: migrated disk to v%d format\r\n", U3D_VER1);
+  }
 
   return;
 }
\ No newline at end of file
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index 334a9dddd6..c4fd9de29c 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -2097,7 +2097,7 @@ _cw_chop(c3_i argc, c3_c* argv[])
     exit(1);
   }
 
-  //  delete all but the last epoch
+  //  delete all but the newly created epoch
   u3_dent* den_u = log_u->com_u->dil_u;
   c3_d epo_d = 0;
   while ( den_u && epo_d <= lat_d ) {
@@ -2191,9 +2191,9 @@ _cw_roll(c3_i argc, c3_c* argv[])
     exit(1);
   }
 
-  //  report success
+  //  success
   c3_d epo_d = log_u->dun_d + 1;
-  fprintf(stderr, "roll: success: created epoch 0i%" PRIu64 "\r\n", epo_d);
+  fprintf(stderr, "roll: epoch rollover complete");
 }
 
 /* _cw_vere(): download vere

From f8834e0871de94fbf2a55458b23f8c6cffb87fc0 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Wed, 26 Apr 2023 13:51:17 -0400
Subject: [PATCH 09/59] disk: validate current snapshot before migration

---
 pkg/vere/disk.c | 116 ++++++++++++++++++++++++------------------------
 pkg/vere/main.c |  10 +++++
 pkg/vere/vere.h |   3 +-
 3 files changed, 70 insertions(+), 59 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 72dd3fcf91..bd6f8a4d27 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -878,31 +878,33 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
   //  XX closures?
   //
   {
-    c3_c* log_c = c3_malloc(10 + strlen(pax_c));
-
-    strcpy(log_c, pax_c);
-    strcat(log_c, "/.urb/log");
+    c3_c log_c[8193];
+    snprintf(log_c, sizeof(log_c), "%s/.urb/log", pax_c);
 
     if ( 0 == (log_u->com_u = u3_foil_folder(log_c)) ) {
       fprintf(stderr, "disk: failed to load /.urb/log in %s\r\n", pax_c);
-      c3_free(log_c);
       c3_free(log_u);
       return 0;
     }
 
-    //  XX validation before migration
-
     //  migrate to the correct disk format if necessary
-    u3_disk_migrate(log_u);
+    if ( c3n == u3_disk_migrate(log_u) ) {
+      fprintf(stderr, "disk: failed to migrate to v%d\r\n", U3D_VER1);
+      c3_free(log_u);
+      return 0;
+    }
 
     //  get latest epoch number
     c3_d lat_d;
-    u3_disk_epoc_last(log_u, &lat_d);
+    if ( c3n == u3_disk_epoc_last(log_u, &lat_d) ) {
+      fprintf(stderr, "disk: failed to load epoch number\r\n");
+      c3_free(log_u);
+      return 0;
+    }
 
     //  initialize epoch's db
     c3_c epo_c[8193];
     snprintf(epo_c, 8192, "%s/0i%" PRIu64, log_c, lat_d);
-    c3_free(log_c);
     {
       // XX extract into function?
       const size_t siz_i =
@@ -932,16 +934,15 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
     if ( 0 == las_d ) {               //  fresh epoch (no events in lmdb yet)
       if ( 0 == lat_d ) {             //  first epoch
         log_u->dun_d = 0;
-        log_u->sen_d = 0;
       } else {                        //  not first epoch
         log_u->dun_d = lat_d - 1;     //  set dun_d to last event in prev epoch
-        log_u->sen_d = log_u->dun_d;
       }
     } else {                          //  not fresh epoch            
       log_u->dun_d = las_d;           //  set dun_d to last event in lmdb        
-      log_u->sen_d = las_d;
     }
 
+    log_u->sen_d = log_u->dun_d;
+
     //  mark the log as live
     log_u->liv_o = c3y;
   }
@@ -992,6 +993,9 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
   fclose(epv_f);
 
   //  create binary version file
+  //  XX make sure if you've updated your binary, a new epoch is created
+  //  so that events within an epoch are guaranteed to have been processed
+  //  by the binary version specified in the file
   c3_c biv_c[8193];
   snprintf(biv_c, sizeof(biv_c), "%s/vere.txt", epo_c);
   FILE* biv_f = fopen(biv_c, "a");
@@ -1038,7 +1042,7 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
   if ( c3y == eps_o ) {
     if ( c3n == u3_disk_save_meta(log_u->mdb_u, who_d, fak_o, lif_w) ) {
       fprintf(stderr, "disk: failed to save metadata\r\n");
-      exit(1);
+      goto fail;
     }
   }
 
@@ -1093,9 +1097,7 @@ c3_o u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d) {
   u3_dent* den_u = u3_foil_folder(log_u->com_u->pax_c)->dil_u;
   while ( den_u ) {
     c3_d epo_d = 0;
-    if ( 1 != sscanf(den_u->nam_c, "0i%" PRIu64, &epo_d) ) {
-      fprintf(stderr, "disk: epoch directory is not a @ui: %s\r\n", den_u->nam_c);
-    } else {
+    if ( 1 == sscanf(den_u->nam_c, "0i%" PRIu64, &epo_d) ) {
       ret_d = c3y;   //  return yes if at least one epoch directory exists
     }
     *lat_d = c3_max(epo_d, *lat_d);  //  update the latest epoch number
@@ -1107,7 +1109,7 @@ c3_o u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d) {
 
 /* u3_disk_migrate: migrates disk format.
  */
-void u3_disk_migrate(u3_disk* log_u)
+c3_o u3_disk_migrate(u3_disk* log_u)
 {
   //  check if data.mdb exists in log directory
   c3_o dat_o = c3n;
@@ -1128,12 +1130,9 @@ void u3_disk_migrate(u3_disk* log_u)
     //  initialize first epoch "0i0"
     if ( c3n == u3_disk_epoc_init(log_u) ) {
       fprintf(stderr, "disk: migrate: failed to initialize first epoch\r\n");
-      exit(1);
+      return c3n;
     }
   } else if ( c3n == epo_o ) {
-    //  migrate existing, unversioned disk to v1
-    fprintf(stderr, "disk: migrating disk to v%d format\r\n", U3D_VER1);
-
     //  initialize pre-migrated lmdb
     MDB_env* old_u;
     {
@@ -1147,8 +1146,7 @@ void u3_disk_migrate(u3_disk* log_u)
 
       if ( 0 == (old_u = u3_lmdb_init(log_u->com_u->pax_c, siz_i)) ) {
         fprintf(stderr, "disk: failed to initialize database\r\n");
-        c3_free(log_u);
-        return;
+        return c3n;
       }
     }
 
@@ -1156,69 +1154,64 @@ void u3_disk_migrate(u3_disk* log_u)
     c3_d fir_d, las_d;
     if ( c3n == u3_lmdb_gulf(old_u, &fir_d, &las_d) ) {
       fprintf(stderr, "disk: failed to get first/last event numbers\r\n");
-      return;
+      return c3n;
     }
 
-    //  boot
-    u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
-
-    //  make sure there's a current snapshot
-    if ( las_d != u3A->eve_d ) {
+    // ensure there's a current snapshot without the below code
+    if ( u3_Host.eve_d != las_d ) {
       fprintf(stderr, "disk: migrate: error: snapshot is out of date, please "
                       "start/shutdown your pier gracefully first\r\n");
-      exit(1);
+      return c3n;
     }
 
-    //  shutdown pre-migrated u3 system and lmdb
-    u3m_stop();
-
     //  initialize first epoch "0i0"
     if ( c3n == u3_disk_epoc_init(log_u) ) {
       fprintf(stderr, "disk: migrate: failed to initialize first epoch\r\n");
-      exit(1);
+      return c3n;
     }
 
-    //  move data.mdb to 0i0/data.mdb
-    c3_c new_c[8193];
-    snprintf(new_c, sizeof(new_c), "%s/0i0/data.mdb", log_u->com_u->pax_c);
-    if ( 0 != rename(dat_c, new_c) ) {
+    //  move data.mdb and lock.mdb to 0i0/
+    c3_c dut_c[8193], luk_c[8193];
+    snprintf(dut_c, sizeof(dut_c), "%s/data.mdb", log_u->com_u->pax_c);
+    snprintf(luk_c, sizeof(luk_c), "%s/lock.mdb", log_u->com_u->pax_c);
+
+    c3_c epo_c[8193], dat_c[8193], lok_c[8193];
+    snprintf(epo_c, sizeof(epo_c), "%s/0i0", log_u->com_u->pax_c);
+    snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", epo_c);
+    snprintf(lok_c, sizeof(lok_c), "%s/lock.mdb", epo_c);
+
+    if ( 0 != c3_rename(dut_c, dat_c) ) {
       fprintf(stderr, "disk: migrate: failed to move data.mdb\r\n");
-      exit(1);
+      goto _u3_disk_migrate_fail;
     }
-
-    //  move lock.mdb to 0i0/lock.mdb
-    c3_c old_c[8193];
-    snprintf(old_c, sizeof(old_c), "%s/lock.mdb", log_u->com_u->pax_c);
-    snprintf(new_c, sizeof(new_c), "%s/0i0/lock.mdb", log_u->com_u->pax_c);
-    if ( 0 != rename(old_c, new_c) ) {
+    if ( 0 != c3_rename(luk_c, lok_c) ) {
       fprintf(stderr, "disk: migrate: failed to move lock.mdb\r\n");
-      exit(1);
+      c3_rename(dat_c, dut_c);
+      goto _u3_disk_migrate_fail;
     }
 
     //  rollover to new epoch
     log_u->mdb_u = old_u;
     if ( c3n == u3_disk_epoc_init(log_u) ) {
       fprintf(stderr, "roll: error: failed to initialize new epoch\r\n");
-      exit(1);
+      goto _u3_disk_migrate_fail;
     }
 
-    //  delete backup snapshot
+    //  delete backup snapshot; migration still succeeds if this fails
     c3_c bhk_c[8193];
     snprintf(bhk_c, sizeof(bhk_c), "%s/.urb/bhk", u3_Host.dir_c);
-    c3_c nop_c[8193];
+
+    c3_c nop_c[8193], sop_c[8193];
     snprintf(nop_c, sizeof(nop_c), "%s/north.bin", bhk_c);
-    c3_c sop_c[8193];
     snprintf(sop_c, sizeof(sop_c), "%s/south.bin", bhk_c);
+
     if ( c3n == c3_unlink(nop_c) ) {
-      fprintf(stderr, "disk: migrate: failed to delete north.bin\r\n");
-      exit(1);
+      fprintf(stderr, "disk: migrate: failed to delete bhk/north.bin\r\n");
     } else if ( c3n == c3_unlink(sop_c) ) {
-      fprintf(stderr, "disk: migrate: failed to delete south.bin\r\n");
-      exit(1);
+      fprintf(stderr, "disk: migrate: failed to delete bhk/south.bin\r\n");
     } else {
       if ( c3n == c3_rmdir(bhk_c) ) {
-        fprintf(stderr, "disk: migrate: failed to delete bhk\r\n");
-        exit(1);
+        fprintf(stderr, "disk: migrate: failed to delete bhk/\r\n");
       }
     }
 
@@ -1226,5 +1219,12 @@ void u3_disk_migrate(u3_disk* log_u)
     fprintf(stderr, "disk: migrated disk to v%d format\r\n", U3D_VER1);
   }
 
-  return;
+_u3_disk_migrate_fail:
+{
+  c3_d lat_d;
+  u3_disk_epoc_last(log_u, &lat_d);
+  u3_disk_epoc_kill(log_u, lat_d);
+}
+
+  return c3y;
 }
\ No newline at end of file
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index c4fd9de29c..3dbe724ea7 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -1971,6 +1971,8 @@ _cw_play(c3_i argc, c3_c* argv[])
 static void
 _cw_prep(c3_i argc, c3_c* argv[])
 {
+  //  XX roll with old binary
+  //     check that new epoch is empty, migrate snapshot in-place
   c3_i ch_i, lid_i;
   c3_w arg_w;
 
@@ -2025,6 +2027,7 @@ _cw_prep(c3_i argc, c3_c* argv[])
 static void
 _cw_chop(c3_i argc, c3_c* argv[])
 {
+  //  XX keep the last epoch (2 epochs left after chop)
   c3_i ch_i, lid_i;
   c3_w arg_w;
 
@@ -2630,6 +2633,13 @@ main(c3_i   argc,
       }
     }
 
+    //  we need the current snapshot's latest event number to 
+    //  validate whether we can execute disk migration
+    if ( c3n == u3_Host.ops_u.nuu ) {
+      u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
+      //  XX u3e_close()
+    }
+
     //  starting u3m configures OpenSSL memory functions, so we must do it
     //  before any OpenSSL allocations
     //
diff --git a/pkg/vere/vere.h b/pkg/vere/vere.h
index a2726d2341..eeaa8cec4f 100644
--- a/pkg/vere/vere.h
+++ b/pkg/vere/vere.h
@@ -312,6 +312,7 @@
       typedef struct _u3_host {
         c3_w       kno_w;                   //  current executing stage
         c3_c*      dir_c;                   //  pier path (no trailing /)
+        c3_d       eve_d;                   //  initial current snapshot
         c3_c*      dem_c;                   //  daemon executable path
         c3_c*      wrk_c;                   //  worker executable path
         c3_d       now_d;                   //  event tick
@@ -999,7 +1000,7 @@
       
       /* u3_disk_migrate(): migrates disk format.
        */
-        void
+        c3_o
         u3_disk_migrate(u3_disk* log_u);
 
       /* u3_lord_init(): start serf.

From b93508e1704090ad2796b2e3c8a4071ab68b5fc8 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Fri, 28 Apr 2023 11:29:32 -0400
Subject: [PATCH 10/59] lmdb: factor siz_i calculation

---
 pkg/vere/db/lmdb.c | 11 ++++++++++-
 pkg/vere/db/lmdb.h |  2 +-
 pkg/vere/disk.c    | 32 ++++----------------------------
 3 files changed, 15 insertions(+), 30 deletions(-)

diff --git a/pkg/vere/db/lmdb.c b/pkg/vere/db/lmdb.c
index 8f8864b77f..d9fed4c0de 100644
--- a/pkg/vere/db/lmdb.c
+++ b/pkg/vere/db/lmdb.c
@@ -38,11 +38,20 @@ intmax_t mdb_get_filesize(mdb_filehandle_t han_u);
 /* u3_lmdb_init(): open lmdb at [pax_c], mmap up to [siz_i].
 */
 MDB_env*
-u3_lmdb_init(const c3_c* pax_c, size_t siz_i)
+u3_lmdb_init(const c3_c* pax_c)
 {
   MDB_env* env_u;
   c3_w     ret_w;
 
+  //  calculate db size based on architecture
+  const size_t siz_i =
+  // 500 GiB is as large as musl on aarch64 wants to allow
+  #if (defined(U3_CPU_aarch64) && defined(U3_OS_linux))
+    0x7d00000000;
+  #else
+    0x10000000000;
+  #endif
+
   if ( (ret_w = mdb_env_create(&env_u)) ) {
     mdb_logerror(stderr, ret_w, "lmdb: init fail");
     return 0;
diff --git a/pkg/vere/db/lmdb.h b/pkg/vere/db/lmdb.h
index aab4e6c9f7..7aa996001f 100644
--- a/pkg/vere/db/lmdb.h
+++ b/pkg/vere/db/lmdb.h
@@ -22,7 +22,7 @@
     /* u3_lmdb_init(): open lmdb at [pax_c], mmap up to [siz_i].
     */
       MDB_env*
-      u3_lmdb_init(const c3_c* pax_c, size_t siz_i);
+      u3_lmdb_init(const c3_c* pax_c);
 
     /* u3_lmdb_exit(): close lmdb.
     */
diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index bd6f8a4d27..19dce9087c 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -906,16 +906,8 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
     c3_c epo_c[8193];
     snprintf(epo_c, 8192, "%s/0i%" PRIu64, log_c, lat_d);
     {
-      // XX extract into function?
-      const size_t siz_i =
-      // 500 GiB is as large as musl on aarch64 wants to allow
-      #if (defined(U3_CPU_aarch64) && defined(U3_OS_linux))
-        0x7d00000000;
-      #else
-        0x10000000000;
-      #endif
-
-      if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c, siz_i)) ) {
+
+      if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c)) ) {
         fprintf(stderr, "disk: failed to initialize database\r\n");
         c3_free(log_u);
         return 0;
@@ -1023,15 +1015,7 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
 
   //  initialize db of new epoch
   {
-    const size_t siz_i =
-    // 500 GiB is as large as musl on aarch64 wants to allow
-    #if (defined(U3_CPU_aarch64) && defined(U3_OS_linux))
-      0x7d00000000;
-    #else
-      0x10000000000;
-    #endif
-
-    if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c, siz_i)) ) {
+    if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c)) ) {
       fprintf(stderr, "disk: failed to initialize database\r\n");
       c3_free(log_u);
       goto fail;
@@ -1136,15 +1120,7 @@ c3_o u3_disk_migrate(u3_disk* log_u)
     //  initialize pre-migrated lmdb
     MDB_env* old_u;
     {
-      const size_t siz_i =
-      // 500 GiB is as large as musl on aarch64 wants to allow
-      #if (defined(U3_CPU_aarch64) && defined(U3_OS_linux))
-        0x7d00000000;
-      #else
-        0x10000000000;
-      #endif
-
-      if ( 0 == (old_u = u3_lmdb_init(log_u->com_u->pax_c, siz_i)) ) {
+      if ( 0 == (old_u = u3_lmdb_init(log_u->com_u->pax_c)) ) {
         fprintf(stderr, "disk: failed to initialize database\r\n");
         return c3n;
       }

From bb1f786c2a72abe6e111262e0b8c435d79defa2c Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Fri, 28 Apr 2023 13:04:46 -0400
Subject: [PATCH 11/59] disk: create new epoch when vere version is different

---
 pkg/vere/disk.c | 69 +++++++++++++++++++++++++++++++++++++------------
 pkg/vere/main.c | 24 ++++++++++++++---
 2 files changed, 74 insertions(+), 19 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 19dce9087c..2c1d1e0622 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -902,9 +902,48 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
       return 0;
     }
 
-    //  initialize epoch's db
+    //  set path to latest epoch
     c3_c epo_c[8193];
     snprintf(epo_c, 8192, "%s/0i%" PRIu64, log_c, lat_d);
+
+    //  get vere version from the latest epoch
+    c3_c fil_c[8193];
+    snprintf(fil_c, 8192, "%s/vere.txt", epo_c);
+    if ( 0 != access(fil_c, F_OK) ) {
+      fprintf(stderr, "disk: failed to access %s\r\n", fil_c);
+      c3_free(log_u);
+      return 0;
+    }
+
+    //  read string from vere.txt file
+    c3_c ver_c[8193];
+    {
+      FILE* fil_u = fopen(fil_c, "r");
+      if ( !fil_u ) {
+        fprintf(stderr, "disk: failed to open %s\r\n", fil_c);
+        c3_free(log_u);
+        return 0;
+      }
+      if ( !fgets(ver_c, 8192, fil_u) ) {
+        fprintf(stderr, "disk: failed to read %s\r\n", fil_c);
+        c3_free(log_u);
+        return 0;
+      }
+      fclose(fil_u);
+    }
+
+    //  if binary version from vere.txt is different 
+    //  than current version, then create a new epoch
+    if ( 0 != strcmp(ver_c, URBIT_VERSION) ) {
+      fprintf(stderr, "disk: binary version mismatch, creating new epoch\r\n");
+      if ( c3n == u3_disk_epoc_init(log_u) ) {
+        fprintf(stderr, "disk: failed to initialize epoch\r\n");
+        c3_free(log_u);
+        return 0;
+      }
+    }
+
+    //  initialize epoch's db
     {
 
       if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c)) ) {
@@ -956,16 +995,12 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
   if ( c3n == eps_o ) {
     new_d = 0;  //  no epochs yet, so create first one 0i0
   } else {
-    //  ensure latest epoch isn't empty
+    //  get first/last event numbers
     c3_d fir_d, las_d;
     if ( c3n == u3_lmdb_gulf(log_u->mdb_u, &fir_d, &las_d) ) {
       fprintf(stderr, "disk: failed to get first/last event numbers\r\n");
       return c3n;
     }
-    if ( fir_d == las_d ) {
-      fprintf(stderr, "disk: latest epoch is empty; skipping rollover\r\n");
-      return c3n;
-    }
     new_d = 1 + las_d;  //  create next epoch
   }
 
@@ -1117,6 +1152,9 @@ c3_o u3_disk_migrate(u3_disk* log_u)
       return c3n;
     }
   } else if ( c3n == epo_o ) {
+    //  migrate existing pier
+    fprintf(stderr, "disk: migrating disk to v%d format\r\n", U3D_VER1);
+
     //  initialize pre-migrated lmdb
     MDB_env* old_u;
     {
@@ -1158,19 +1196,25 @@ c3_o u3_disk_migrate(u3_disk* log_u)
 
     if ( 0 != c3_rename(dut_c, dat_c) ) {
       fprintf(stderr, "disk: migrate: failed to move data.mdb\r\n");
-      goto _u3_disk_migrate_fail;
+      u3_disk_epoc_last(log_u, &lat_d);
+      u3_disk_epoc_kill(log_u, lat_d);
+      return c3n;
     }
     if ( 0 != c3_rename(luk_c, lok_c) ) {
       fprintf(stderr, "disk: migrate: failed to move lock.mdb\r\n");
       c3_rename(dat_c, dut_c);
-      goto _u3_disk_migrate_fail;
+      u3_disk_epoc_last(log_u, &lat_d);
+      u3_disk_epoc_kill(log_u, lat_d);
+      return c3n;
     }
 
     //  rollover to new epoch
     log_u->mdb_u = old_u;
     if ( c3n == u3_disk_epoc_init(log_u) ) {
       fprintf(stderr, "roll: error: failed to initialize new epoch\r\n");
-      goto _u3_disk_migrate_fail;
+      u3_disk_epoc_last(log_u, &lat_d);
+      u3_disk_epoc_kill(log_u, lat_d);
+      return c3n;
     }
 
     //  delete backup snapshot; migration still succeeds if this fails
@@ -1195,12 +1239,5 @@ c3_o u3_disk_migrate(u3_disk* log_u)
     fprintf(stderr, "disk: migrated disk to v%d format\r\n", U3D_VER1);
   }
 
-_u3_disk_migrate_fail:
-{
-  c3_d lat_d;
-  u3_disk_epoc_last(log_u, &lat_d);
-  u3_disk_epoc_kill(log_u, lat_d);
-}
-
   return c3y;
 }
\ No newline at end of file
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index 3dbe724ea7..ba073d1b21 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -2088,7 +2088,16 @@ _cw_chop(c3_i argc, c3_c* argv[])
   }
 
   //  create new epoch
-  if ( c3n == u3_disk_epoc_init(log_u) ) {
+  c3_d fir_d, las_d;
+  if ( c3n == u3_lmdb_gulf(log_u->mdb_u, &fir_d, &las_d) ) {
+    fprintf(stderr, "chop: failed to get first/last events\r\n");
+    exit(1);
+  }
+  if ( fir_d == las_d == 0 ) {
+    fprintf(stderr, "chop: latest epoch already empty\r\n");
+    exit(1);
+  }
+  else if ( c3n == u3_disk_epoc_init(log_u) ) {
     fprintf(stderr, "chop: failed to create new epoch\r\n");
     exit(1);
   }
@@ -2189,8 +2198,17 @@ _cw_roll(c3_i argc, c3_c* argv[])
   }
 
   //  create new epoch
-  if ( c3n == u3_disk_epoc_init(log_u) ) {
-    fprintf(stderr, "roll: error: failed to initialize new epoch\r\n");
+  c3_d fir_d, las_d;
+  if ( c3n == u3_lmdb_gulf(log_u->mdb_u, &fir_d, &las_d) ) {
+    fprintf(stderr, "roll: failed to get first/last events\r\n");
+    exit(1);
+  }
+  if ( fir_d == las_d == 0 ) {
+    fprintf(stderr, "roll: latest epoch already empty\r\n");
+    exit(1);
+  }
+  else if ( c3n == u3_disk_epoc_init(log_u) ) {
+    fprintf(stderr, "roll: failed to create new epoch\r\n");
     exit(1);
   }
 

From 5b432b6b3ec8188d002ff8a32f0c52522dc0c1e9 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Mon, 1 May 2023 08:54:08 -0400
Subject: [PATCH 12/59] cli: make `chop` leave only the latest two epochs

---
 pkg/vere/disk.c | 42 ++----------------------------------------
 pkg/vere/main.c | 25 ++++++++++++++++---------
 2 files changed, 18 insertions(+), 49 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 2c1d1e0622..d4ca5a21b8 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -906,43 +906,6 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
     c3_c epo_c[8193];
     snprintf(epo_c, 8192, "%s/0i%" PRIu64, log_c, lat_d);
 
-    //  get vere version from the latest epoch
-    c3_c fil_c[8193];
-    snprintf(fil_c, 8192, "%s/vere.txt", epo_c);
-    if ( 0 != access(fil_c, F_OK) ) {
-      fprintf(stderr, "disk: failed to access %s\r\n", fil_c);
-      c3_free(log_u);
-      return 0;
-    }
-
-    //  read string from vere.txt file
-    c3_c ver_c[8193];
-    {
-      FILE* fil_u = fopen(fil_c, "r");
-      if ( !fil_u ) {
-        fprintf(stderr, "disk: failed to open %s\r\n", fil_c);
-        c3_free(log_u);
-        return 0;
-      }
-      if ( !fgets(ver_c, 8192, fil_u) ) {
-        fprintf(stderr, "disk: failed to read %s\r\n", fil_c);
-        c3_free(log_u);
-        return 0;
-      }
-      fclose(fil_u);
-    }
-
-    //  if binary version from vere.txt is different 
-    //  than current version, then create a new epoch
-    if ( 0 != strcmp(ver_c, URBIT_VERSION) ) {
-      fprintf(stderr, "disk: binary version mismatch, creating new epoch\r\n");
-      if ( c3n == u3_disk_epoc_init(log_u) ) {
-        fprintf(stderr, "disk: failed to initialize epoch\r\n");
-        c3_free(log_u);
-        return 0;
-      }
-    }
-
     //  initialize epoch's db
     {
 
@@ -966,12 +929,11 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
       if ( 0 == lat_d ) {             //  first epoch
         log_u->dun_d = 0;
       } else {                        //  not first epoch
-        log_u->dun_d = lat_d - 1;     //  set dun_d to last event in prev epoch
+        log_u->dun_d = lat_d;         //  set dun_d to last event in prev epoch
       }
     } else {                          //  not fresh epoch            
       log_u->dun_d = las_d;           //  set dun_d to last event in lmdb        
     }
-
     log_u->sen_d = log_u->dun_d;
 
     //  mark the log as live
@@ -1001,7 +963,7 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
       fprintf(stderr, "disk: failed to get first/last event numbers\r\n");
       return c3n;
     }
-    new_d = 1 + las_d;  //  create next epoch
+    new_d = las_d;  //  create next epoch
   }
 
   //  create new epoch directory
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index ba073d1b21..b122aa45ab 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -2027,7 +2027,6 @@ _cw_prep(c3_i argc, c3_c* argv[])
 static void
 _cw_chop(c3_i argc, c3_c* argv[])
 {
-  //  XX keep the last epoch (2 epochs left after chop)
   c3_i ch_i, lid_i;
   c3_w arg_w;
 
@@ -2087,13 +2086,20 @@ _cw_chop(c3_i argc, c3_c* argv[])
     exit(1);
   }
 
+  //  get latest epoch number prior to creating a new one
+  c3_d pre_d;
+  if ( c3n == u3_disk_epoc_last(log_u, &pre_d) ) {
+    fprintf(stderr, "chop: failed to find last epoch\r\n");
+    exit(1);
+  }
+
   //  create new epoch
   c3_d fir_d, las_d;
   if ( c3n == u3_lmdb_gulf(log_u->mdb_u, &fir_d, &las_d) ) {
     fprintf(stderr, "chop: failed to get first/last events\r\n");
     exit(1);
   }
-  if ( fir_d == las_d == 0 ) {
+  if ( fir_d == las_d ) {
     fprintf(stderr, "chop: latest epoch already empty\r\n");
     exit(1);
   }
@@ -2102,20 +2108,20 @@ _cw_chop(c3_i argc, c3_c* argv[])
     exit(1);
   }
 
-  //  get latest epoch number
-  c3_d lat_d;
-  if ( c3n == u3_disk_epoc_last(log_u, &lat_d) ) {
+  //  get latest epoch number prior to creating a new one
+  c3_d pos_d;
+  if ( c3n == u3_disk_epoc_last(log_u, &pos_d) ) {
     fprintf(stderr, "chop: failed to find last epoch\r\n");
     exit(1);
   }
 
-  //  delete all but the newly created epoch
+  //  delete all but the last two epochs
   u3_dent* den_u = log_u->com_u->dil_u;
   c3_d epo_d = 0;
-  while ( den_u && epo_d <= lat_d ) {
+  while ( den_u ) {
     if ( 1 != sscanf(den_u->nam_c, "0i%" PRIu64, &epo_d) ) {
       fprintf(stderr, "disk: epoch directory is not a @ui: %s\r\n", den_u->nam_c);
-    } else {
+    } else if ( epo_d != pre_d && epo_d != pos_d ) {
       fprintf(stderr, "chop: deleting epoch 0i%" PRIu64 "\r\n", epo_d);
       if ( c3y != u3_disk_epoc_kill(log_u, epo_d) ) {
         fprintf(stderr, "chop: failed to delete epoch 0i%" PRIu64 "\r\n", epo_d);
@@ -2203,7 +2209,8 @@ _cw_roll(c3_i argc, c3_c* argv[])
     fprintf(stderr, "roll: failed to get first/last events\r\n");
     exit(1);
   }
-  if ( fir_d == las_d == 0 ) {
+
+  if ( fir_d == las_d ) {
     fprintf(stderr, "roll: latest epoch already empty\r\n");
     exit(1);
   }

From 2e9da020793173c32d3589330f7b148f11530f1c Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Mon, 1 May 2023 10:16:10 -0400
Subject: [PATCH 13/59] disk: create a new epoch when version change detected

---
 pkg/vere/disk.c | 56 ++++++++++++++++++++++++++++++++++++++++---------
 pkg/vere/main.c |  5 ++---
 pkg/vere/vere.h |  5 +++++
 3 files changed, 53 insertions(+), 13 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index d4ca5a21b8..866e6b9391 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -875,7 +875,6 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
   }
 
   //  create/load $pier/.urb/log
-  //  XX closures?
   //
   {
     c3_c log_c[8193];
@@ -902,18 +901,23 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
       return 0;
     }
 
+    //  get binary version from latest epoch
+    c3_c ver_w[8193];
+    if ( c3n == u3_disk_epoc_vere(log_u, lat_d, ver_w) ) {
+      fprintf(stderr, "disk: failed to load epoch version\r\n");
+      c3_free(log_u);
+      return 0;
+    }
+
     //  set path to latest epoch
     c3_c epo_c[8193];
     snprintf(epo_c, 8192, "%s/0i%" PRIu64, log_c, lat_d);
 
-    //  initialize epoch's db
-    {
-
-      if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c)) ) {
-        fprintf(stderr, "disk: failed to initialize database\r\n");
-        c3_free(log_u);
-        return 0;
-      }
+    //  initialize latest epoch's db
+    if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c)) ) {
+      fprintf(stderr, "disk: failed to initialize database\r\n");
+      c3_free(log_u);
+      return 0;
     }
     fprintf(stderr, "disk: loaded epoch 0i%" PRIu64 "\r\n", lat_d);
 
@@ -936,6 +940,16 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
     }
     log_u->sen_d = log_u->dun_d;
 
+    //  if binary version of latest epoch is not the same as the
+    //  running binary, then we need to create a new epoch
+    if ( 0 != strcmp(ver_w, URBIT_VERSION) ) {
+      if ( c3n == u3_disk_epoc_init(log_u) ) {
+        fprintf(stderr, "disk: failed to initialize epoch\r\n");
+        c3_free(log_u);
+        return 0;
+      }
+    }
+
     //  mark the log as live
     log_u->liv_o = c3y;
   }
@@ -1088,6 +1102,29 @@ c3_o u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d) {
   return ret_d;
 }
 
+/* u3_disk_epoc_vere: get binary version from epoch.
+*/
+c3_o
+u3_disk_epoc_vere(u3_disk* log_u, c3_d epo_d, c3_c* ver_w) {
+  c3_c ver_c[8193];
+  snprintf(ver_c, sizeof(ver_c), "%s/0i%" PRIu64 "/vere.txt", 
+                  log_u->com_u->pax_c, epo_d);
+
+  FILE* fil_u = fopen(ver_c, "r");
+  if ( NULL == fil_u ) {
+    fprintf(stderr, "disk: failed to open vere.txt in epoch 0i%" PRIu64 
+                    "\r\n", epo_d);
+    return c3n;
+  }
+
+  if ( 1 != fscanf(fil_u, "%s", ver_w) ) {
+    fprintf(stderr, "disk: failed to read vere.txt in epoch 0i%" PRIu64 
+                    "\r\n", epo_d);
+    return c3n;
+  }
+  return c3y;
+}
+
 /* u3_disk_migrate: migrates disk format.
  */
 c3_o u3_disk_migrate(u3_disk* log_u)
@@ -1115,7 +1152,6 @@ c3_o u3_disk_migrate(u3_disk* log_u)
     }
   } else if ( c3n == epo_o ) {
     //  migrate existing pier
-    fprintf(stderr, "disk: migrating disk to v%d format\r\n", U3D_VER1);
 
     //  initialize pre-migrated lmdb
     MDB_env* old_u;
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index b122aa45ab..838cac5388 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -2119,9 +2119,8 @@ _cw_chop(c3_i argc, c3_c* argv[])
   u3_dent* den_u = log_u->com_u->dil_u;
   c3_d epo_d = 0;
   while ( den_u ) {
-    if ( 1 != sscanf(den_u->nam_c, "0i%" PRIu64, &epo_d) ) {
-      fprintf(stderr, "disk: epoch directory is not a @ui: %s\r\n", den_u->nam_c);
-    } else if ( epo_d != pre_d && epo_d != pos_d ) {
+    c3_d res_d = sscanf(den_u->nam_c, "0i%" PRIu64, &epo_d);
+    if ( (1 == res_d) && (epo_d != pre_d) && (epo_d != pos_d) ) {
       fprintf(stderr, "chop: deleting epoch 0i%" PRIu64 "\r\n", epo_d);
       if ( c3y != u3_disk_epoc_kill(log_u, epo_d) ) {
         fprintf(stderr, "chop: failed to delete epoch 0i%" PRIu64 "\r\n", epo_d);
diff --git a/pkg/vere/vere.h b/pkg/vere/vere.h
index eeaa8cec4f..f54fa2a6d6 100644
--- a/pkg/vere/vere.h
+++ b/pkg/vere/vere.h
@@ -998,6 +998,11 @@
        c3_o
        u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d);
       
+      /* u3_disk_epoc_vere(): get binary version from epoch.
+       */
+       c3_o
+       u3_disk_epoc_vere(u3_disk* log_u, c3_d epo_d, c3_c* ver_w);
+      
       /* u3_disk_migrate(): migrates disk format.
        */
         c3_o

From 1b952d06aecc24dde76f9944466c71149fd905d2 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Tue, 16 May 2023 12:02:01 -0400
Subject: [PATCH 14/59] c3: add `c3_link`

---
 pkg/c3/defs.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/pkg/c3/defs.h b/pkg/c3/defs.h
index 38b39263a9..bbf8d91ef7 100644
--- a/pkg/c3/defs.h
+++ b/pkg/c3/defs.h
@@ -149,6 +149,8 @@
         mkdir(a, b);})
 #     define c3_rmdir(a) ({                                     \
         rmdir(a);})
+#     define c3_link(a, b) ({                                   \
+        link(a, b);})
 #     define c3_unlink(a) ({                                    \
         unlink(a);})
 #     define c3_fopen(a, b) ({                                  \

From b0da9beb58ee47ac306fe03d01e8ac461156ce13 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Thu, 18 May 2023 13:37:08 -0400
Subject: [PATCH 15/59] disk: make migration idempotent (crash safe)

---
 pkg/vere/disk.c | 212 +++++++++++++++++++++++++++++++++++++-----------
 pkg/vere/main.c |   4 +-
 pkg/vere/vere.h |   5 ++
 3 files changed, 173 insertions(+), 48 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 866e6b9391..3985528493 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -962,6 +962,87 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
   return log_u;
 }
 
+/* u3_disk_epoc_good: check for valid epoch.
+*/
+c3_o u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d) {
+  /*  a "valid" epoch is currently defined as a writable folder in 
+   *  the <pier>/.urb/log directory named with a @ui and contains:
+   *  - a writable data.mdb file which can be opened by lmdb (add db checking?)
+   *  - a writable epoc.txt file
+   *  - a writable vere.txt file
+   *
+   *  XX: should we check if the correct event sequence exists in the data.mdb file?
+   *
+   *  note that this does not check if the epoch even contains snapshot files, 
+   *  let alone whether they're valid or not
+   */
+
+  c3_o ret_o = c3y;  //  return code
+
+  //  file paths
+  c3_c epo_c[8193], dat_c[8193], epv_c[8193], biv_c[8193];
+  snprintf(epo_c, sizeof(epo_c), "%s/0i%" PRIu64, log_u->com_u->pax_c, epo_d);
+  snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", epo_c);
+  snprintf(epv_c, sizeof(epv_c), "%s/epoc.txt", epo_c);
+  snprintf(biv_c, sizeof(biv_c), "%s/vere.txt", epo_c);
+
+  c3_o dir_o = c3n;  //  directory is writable
+  c3_o dat_o = c3n;  //  data.mdb is writable
+  c3_o mdb_o = c3n;  //  data.mdb can be opened
+  c3_o epv_o = c3n;  //  epoc.txt is writable
+  c3_o biv_o = c3n;  //  vere.txt is writable
+
+  //  check if dir is writable
+  if ( 0 == access(epo_c, W_OK) ) {
+    dir_o = c3y;
+  }
+
+  //  check if data.mdb is writable
+  if ( 0 == access(dat_c, W_OK) ) {
+    dat_o = c3y;
+    //  check if we can open data.mdb
+    MDB_env* env_u;
+    if ( 0 != (env_u = u3_lmdb_init(epo_c)) ) {
+      mdb_o = c3y;
+    }
+    u3_lmdb_exit(env_u);
+  }
+
+  //  check if epoc.txt is writable
+  if ( 0 == access(epv_c, W_OK) ) {
+    epv_o = c3y;
+  }
+
+  //  check if vere.txt is writable
+  if ( 0 == access(biv_c, W_OK) ) {
+    biv_o = c3y;
+  }
+
+  //  print error messages
+  if ( c3n == dir_o ) {
+    fprintf(stderr, "disk: epoch 0i%" PRIu64 " is not writable\r\n", epo_d);
+    ret_o = c3n;
+  }
+  if ( c3n == dat_o ) {
+    fprintf(stderr, "disk: epoch 0i%" PRIu64 "/data.mdb is not writable\r\n", epo_d);
+    ret_o = c3n;
+  }
+  if ( c3n == mdb_o ) {
+    fprintf(stderr, "disk: epoch 0i%" PRIu64 "/data.mdb can't be opened\r\n", epo_d);
+    ret_o = c3n;
+  }
+  if ( c3n == epv_o ) {
+    fprintf(stderr, "disk: epoch 0i%" PRIu64 "/epoc.txt is not writable\r\n", epo_d);
+    ret_o = c3n;
+  }
+  if ( c3n == biv_o ) {
+    fprintf(stderr, "disk: epoch 0i%" PRIu64 "/vere.txt is not writable\r\n", epo_d);
+    ret_o = c3n;
+  }
+
+  return ret_o;
+}
+ 
 /* u3_disk_epoc_init: create new epoch.
 */
 c3_o u3_disk_epoc_init(u3_disk* log_u) {
@@ -969,8 +1050,14 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
   c3_d lat_d, new_d;
   c3_o eps_o = u3_disk_epoc_last(log_u, &lat_d);
   if ( c3n == eps_o ) {
-    new_d = 0;  //  no epochs yet, so create first one 0i0
+    //  no epochs yet, so create first one 0i0
+    new_d = 0;
+  } else if ( c3n == u3_disk_epoc_good(log_u, lat_d) ) {
+    //  last epoch is invalid, so overwrite it
+    new_d = lat_d;
   } else {
+    //  last epoch is valid, so create next one
+
     //  get first/last event numbers
     c3_d fir_d, las_d;
     if ( c3n == u3_lmdb_gulf(log_u->mdb_u, &fir_d, &las_d) ) {
@@ -980,28 +1067,26 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
     new_d = las_d;  //  create next epoch
   }
 
-  //  create new epoch directory
+  //  create new epoch directory if it doesn't exist
   c3_c epo_c[8193];
   snprintf(epo_c, sizeof(epo_c), "%s/0i%" PRIu64, log_u->com_u->pax_c, new_d);
-  if ( 0 != c3_mkdir(epo_c, 0700) ) {
+  c3_d ret_d = c3_mkdir(epo_c, 0700);
+  if ( ( ret_d < 0 ) && ( errno != EEXIST ) ) {
     fprintf(stderr, "disk: failed to create epoch directory %" PRIu64 "\r\n", new_d);
     return c3n;
   }
 
-  //  create epoch version file
+  //  create epoch version file, overwriting any existing file
   c3_c epv_c[8193];
   snprintf(epv_c, sizeof(epv_c), "%s/epoc.txt", epo_c);
-  FILE* epv_f = fopen(epv_c, "a");
+  FILE* epv_f = fopen(epv_c, "w");
   fprintf(epv_f, "%d\n", U3D_VER1);
   fclose(epv_f);
 
-  //  create binary version file
-  //  XX make sure if you've updated your binary, a new epoch is created
-  //  so that events within an epoch are guaranteed to have been processed
-  //  by the binary version specified in the file
+  //  create binary version file, overwriting any existing file
   c3_c biv_c[8193];
   snprintf(biv_c, sizeof(biv_c), "%s/vere.txt", epo_c);
-  FILE* biv_f = fopen(biv_c, "a");
+  FILE* biv_f = fopen(biv_c, "w");
   fprintf(biv_f, URBIT_VERSION);
   fclose(biv_f);
 
@@ -1013,7 +1098,7 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
     }
   }
 
-  //  get metadata from old epoch's db
+  //  get metadata from old epoch or unmigrated event log's db
   c3_d     who_d[2];
   c3_o     fak_o;
   c3_w     lif_w;
@@ -1025,7 +1110,13 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
   }
 
   //  initialize db of new epoch
-  {
+  if ( c3y == u3_Host.ops_u.nuu || new_d > 0 ) {
+    c3_c dat_c[8193];
+    snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", epo_c);
+    // if ( c3n == c3_unlink(dat_c) ) {
+    //   fprintf(stderr, "disk: failed to rm 0i%" PRIu64 "/data.mdb\r\n", new_d);
+    //   goto fail;
+    // };
     if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c)) ) {
       fprintf(stderr, "disk: failed to initialize database\r\n");
       c3_free(log_u);
@@ -1093,7 +1184,7 @@ c3_o u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d) {
   while ( den_u ) {
     c3_d epo_d = 0;
     if ( 1 == sscanf(den_u->nam_c, "0i%" PRIu64, &epo_d) ) {
-      ret_d = c3y;   //  return yes if at least one epoch directory exists
+      ret_d = c3y;   //  NB: returns yes if the directory merely exists
     }
     *lat_d = c3_max(epo_d, *lat_d);  //  update the latest epoch number
     den_u = den_u->nex_u;
@@ -1129,19 +1220,42 @@ u3_disk_epoc_vere(u3_disk* log_u, c3_d epo_d, c3_c* ver_w) {
  */
 c3_o u3_disk_migrate(u3_disk* log_u)
 {
-  //  check if data.mdb exists in log directory
+  /*  migration steps:
+   *  0. detect whether we need to migrate or not
+   *     a. if it's a fresh boot via u3_Host.ops_u.nuu -> skip migration
+   *     b. if data.mdb is readable in log directory -> execute migration
+   *  1. initialize epoch 0i0 (first call to u3_disk_epoc_init())
+   *     a. creates epoch directory
+   *     b. creates epoch version file
+   *     c. creates binary version file
+   *     d. initializes database
+   *     e. reads metadata from old database
+   *     f. writes metadata to new database
+   *     g. loads new epoch directory and sets it in log_u
+   *  2. create hard links to data.mdb and lock.mdb in 0i0/
+   *  3. rollover to new epoch (second call to u3_disk_epoc_init())
+   *     a. same as 1a-g but also copies current snapshot between c/d steps
+   *  4. delete backup snapshot (c3_unlink() and c3_rmdir() calls)
+   *  5. delete old data.mdb and lock.mdb files (c3_unlink() calls)
+   */
+
+  //  check if data.mdb is readable in log directory
   c3_o dat_o = c3n;
   c3_c dat_c[8193];
   snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", log_u->com_u->pax_c);
-  if ( 0 == access(dat_c, F_OK) ) {
+  if ( 0 == access(dat_c, R_OK) ) {
     dat_o = c3y;
   }
 
-  //  check if any epochs exist
-  c3_d lat_d;
-  c3_o epo_o = u3_disk_epoc_last(log_u, &lat_d);
+  //  check if lock.mdb is readable in log directory
+  c3_o lok_o = c3n;
+  c3_c lok_c[8193];
+  snprintf(lok_c, sizeof(dat_c), "%s/data.mdb", log_u->com_u->pax_c);
+  if ( 0 == access(dat_c, R_OK) ) {
+    lok_o = c3y;
+  }
 
-  if ( c3n == epo_o && c3n == dat_o ) {
+  if ( c3y == u3_Host.ops_u.nuu ) {
     //  initialize disk v1 on fresh boots
     fprintf(stderr, "disk: initializing disk with v%d format\r\n", U3D_VER1);
 
@@ -1150,13 +1264,14 @@ c3_o u3_disk_migrate(u3_disk* log_u)
       fprintf(stderr, "disk: migrate: failed to initialize first epoch\r\n");
       return c3n;
     }
-  } else if ( c3n == epo_o ) {
-    //  migrate existing pier
+  } else if ( c3y == dat_o ) {
+    //  migrate existing pier which has either:
+    //  - not started the migration, or 
+    //  - crashed before completing the migration
 
     //  initialize pre-migrated lmdb
-    MDB_env* old_u;
     {
-      if ( 0 == (old_u = u3_lmdb_init(log_u->com_u->pax_c)) ) {
+      if ( 0 == (log_u->mdb_u = u3_lmdb_init(log_u->com_u->pax_c)) ) {
         fprintf(stderr, "disk: failed to initialize database\r\n");
         return c3n;
       }
@@ -1164,15 +1279,17 @@ c3_o u3_disk_migrate(u3_disk* log_u)
 
     //  get first/last event numbers from pre-migrated lmdb
     c3_d fir_d, las_d;
-    if ( c3n == u3_lmdb_gulf(old_u, &fir_d, &las_d) ) {
+    if ( c3n == u3_lmdb_gulf(log_u->mdb_u, &fir_d, &las_d) ) {
       fprintf(stderr, "disk: failed to get first/last event numbers\r\n");
       return c3n;
     }
 
-    // ensure there's a current snapshot without the below code
+    // ensure there's a current snapshot
     if ( u3_Host.eve_d != las_d ) {
       fprintf(stderr, "disk: migrate: error: snapshot is out of date, please "
                       "start/shutdown your pier gracefully first\r\n");
+      fprintf(stderr, "disk: migrate: eve_d (%" PRIu64 ") != las_d (%" PRIu64 ")\r\n",
+                      u3_Host.eve_d, las_d);
       return c3n;
     }
 
@@ -1182,47 +1299,41 @@ c3_o u3_disk_migrate(u3_disk* log_u)
       return c3n;
     }
 
-    //  move data.mdb and lock.mdb to 0i0/
-    c3_c dut_c[8193], luk_c[8193];
+    //  create hard links to data.mdb and lock.mdb in 0i0/
+    c3_c dut_c[8193], luk_c[8193];  //  old paths
     snprintf(dut_c, sizeof(dut_c), "%s/data.mdb", log_u->com_u->pax_c);
     snprintf(luk_c, sizeof(luk_c), "%s/lock.mdb", log_u->com_u->pax_c);
 
-    c3_c epo_c[8193], dat_c[8193], lok_c[8193];
+    c3_c epo_c[8193], dat_c[8193], lok_c[8193];  //  new paths
     snprintf(epo_c, sizeof(epo_c), "%s/0i0", log_u->com_u->pax_c);
     snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", epo_c);
     snprintf(lok_c, sizeof(lok_c), "%s/lock.mdb", epo_c);
 
-    if ( 0 != c3_rename(dut_c, dat_c) ) {
-      fprintf(stderr, "disk: migrate: failed to move data.mdb\r\n");
-      u3_disk_epoc_last(log_u, &lat_d);
-      u3_disk_epoc_kill(log_u, lat_d);
+    if ( 0 < c3_link(dut_c, dat_c) ) {
+      fprintf(stderr, "disk: migrate: failed to create data.mdb hard link\r\n");
+      fprintf(stderr, "errno %d: %s\r\n", errno, strerror(errno));
       return c3n;
     }
-    if ( 0 != c3_rename(luk_c, lok_c) ) {
-      fprintf(stderr, "disk: migrate: failed to move lock.mdb\r\n");
-      c3_rename(dat_c, dut_c);
-      u3_disk_epoc_last(log_u, &lat_d);
-      u3_disk_epoc_kill(log_u, lat_d);
-      return c3n;
+    if ( c3y == lok_o ) {  //  only link lock.mdb if it exists
+      if ( 0 < c3_link(luk_c, lok_c) ) {
+        fprintf(stderr, "disk: migrate: failed to create lock.mdb hard link\r\n");
+        c3_rename(dat_c, dut_c);
+        return c3n;
+      }
     }
 
     //  rollover to new epoch
-    log_u->mdb_u = old_u;
+    // log_u->mdb_u = old_u;
     if ( c3n == u3_disk_epoc_init(log_u) ) {
       fprintf(stderr, "roll: error: failed to initialize new epoch\r\n");
-      u3_disk_epoc_last(log_u, &lat_d);
-      u3_disk_epoc_kill(log_u, lat_d);
       return c3n;
     }
 
-    //  delete backup snapshot; migration still succeeds if this fails
-    c3_c bhk_c[8193];
+    //  delete backup snapshot
+    c3_c bhk_c[8193], nop_c[8193], sop_c[8193];
     snprintf(bhk_c, sizeof(bhk_c), "%s/.urb/bhk", u3_Host.dir_c);
-
-    c3_c nop_c[8193], sop_c[8193];
     snprintf(nop_c, sizeof(nop_c), "%s/north.bin", bhk_c);
     snprintf(sop_c, sizeof(sop_c), "%s/south.bin", bhk_c);
-
     if ( c3n == c3_unlink(nop_c) ) {
       fprintf(stderr, "disk: migrate: failed to delete bhk/north.bin\r\n");
     } else if ( c3n == c3_unlink(sop_c) ) {
@@ -1233,6 +1344,15 @@ c3_o u3_disk_migrate(u3_disk* log_u)
       }
     }
 
+    //  delete old lock.mdb and data.mdb files
+    if ( 0 != c3_unlink(luk_c) ) {
+      fprintf(stderr, "disk: migrate: failed to unlink lock.mdb\r\n");
+    }
+    if ( 0 != c3_unlink(dut_c) ) {
+      fprintf(stderr, "disk: migrate: failed to unlink data.mdb\r\n");
+      return c3n;  //  migration succeeds only if we can unlink data.mdb
+    }
+
     //  success
     fprintf(stderr, "disk: migrated disk to v%d format\r\n", U3D_VER1);
   }
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index 838cac5388..7d69ed179f 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -2220,7 +2220,7 @@ _cw_roll(c3_i argc, c3_c* argv[])
 
   //  success
   c3_d epo_d = log_u->dun_d + 1;
-  fprintf(stderr, "roll: epoch rollover complete");
+  fprintf(stderr, "roll: epoch rollover complete\r\n");
 }
 
 /* _cw_vere(): download vere
@@ -2659,7 +2659,7 @@ main(c3_i   argc,
 
     //  we need the current snapshot's latest event number to 
     //  validate whether we can execute disk migration
-    if ( c3n == u3_Host.ops_u.nuu ) {
+    if ( u3_Host.ops_u.nuu == c3n ) {
       u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
       //  XX u3e_close()
     }
diff --git a/pkg/vere/vere.h b/pkg/vere/vere.h
index f54fa2a6d6..1e76053d10 100644
--- a/pkg/vere/vere.h
+++ b/pkg/vere/vere.h
@@ -982,6 +982,11 @@
       */
         void
         u3_disk_plan(u3_disk* log_u, u3_fact* tac_u);
+
+      /* u3_disk_epoc_good(): check for valid epoch.
+       */
+       c3_o
+       u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d);
       
       /* u3_disk_epoc_init(): create new epoch.
        */

From 3b1ee9a0d2b8c5217c16b7bae563c924a7aae751 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Thu, 25 May 2023 15:08:04 -0400
Subject: [PATCH 16/59] cli: fix `chop` bug

---
 pkg/vere/disk.c |  1 +
 pkg/vere/main.c | 64 +++++++++++++++++++++++++++++++++++++++----------
 2 files changed, 52 insertions(+), 13 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 3985528493..ce6930ea89 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -1224,6 +1224,7 @@ c3_o u3_disk_migrate(u3_disk* log_u)
    *  0. detect whether we need to migrate or not
    *     a. if it's a fresh boot via u3_Host.ops_u.nuu -> skip migration
    *     b. if data.mdb is readable in log directory -> execute migration
+   *        if not -> skip migration
    *  1. initialize epoch 0i0 (first call to u3_disk_epoc_init())
    *     a. creates epoch directory
    *     b. creates epoch version file
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index c3da6be457..7f9d360e9a 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -2157,15 +2157,48 @@ _cw_chop(c3_i argc, c3_c* argv[])
     fprintf(stderr, "chop: failed to get first/last events\r\n");
     exit(1);
   }
-  if ( fir_d == las_d ) {
-    fprintf(stderr, "chop: latest epoch already empty\r\n");
-    exit(1);
-  }
-  else if ( c3n == u3_disk_epoc_init(log_u) ) {
+
+  //  create new epoch if latest isn't empty
+  if ( (fir_d != las_d) && (c3n == u3_disk_epoc_init(log_u)) ) {
     fprintf(stderr, "chop: failed to create new epoch\r\n");
     exit(1);
   }
 
+  //  sort epoch directories in descending order
+  u3_dire* ned_u = u3_foil_folder(log_u->com_u->pax_c);
+  u3_dent* den_u = ned_u->dil_u;
+  c3_z len_z = 0;
+  while ( den_u ) {  //  count epochs
+    len_z++;
+    den_u = den_u->nex_u;
+  }
+  c3_d* sot_d = c3_malloc(len_z * sizeof(c3_d));
+  len_z = 0;
+  den_u = ned_u->dil_u;
+  while ( den_u ) {
+    if ( 1 == sscanf(den_u->nam_c, "0i%" PRIu64, (sot_d + len_z)) ) {
+      len_z++;
+    }
+    den_u = den_u->nex_u;
+  }
+
+  if ( len_z <= 2 ) {
+    fprintf(stderr, "chop: nothing to do, have a great day\r\n");
+    exit(0);  //  enjoy
+  }
+
+  //  sort sot_d naively in descending order
+  c3_d tmp_d;
+  for ( c3_z i_z = 0; i_z < len_z; i_z++ ) {
+    for ( c3_z j_z = i_z + 1; j_z < len_z; j_z++ ) {
+      if ( sot_d[i_z] < sot_d[j_z] ) {
+        tmp_d = sot_d[i_z];
+        sot_d[i_z] = sot_d[j_z];
+        sot_d[j_z] = tmp_d;
+      }
+    }
+  }
+
   //  get latest epoch number prior to creating a new one
   c3_d pos_d;
   if ( c3n == u3_disk_epoc_last(log_u, &pos_d) ) {
@@ -2174,20 +2207,25 @@ _cw_chop(c3_i argc, c3_c* argv[])
   }
 
   //  delete all but the last two epochs
-  u3_dent* den_u = log_u->com_u->dil_u;
+  //  XX parameterize the number of epochs to chop
+  for ( c3_z i_z = 2; i_z < len_z; i_z++ ) {
+    fprintf(stderr, "chop: deleting epoch 0i%" PRIu64 "\r\n", sot_d[i_z]);
+    if ( c3y != u3_disk_epoc_kill(log_u, sot_d[i_z]) ) {
+      fprintf(stderr, "chop: failed to delete epoch 0i%" PRIu64 "\r\n", sot_d[i_z]);
+      exit(1);
+    }
+  }
+
   c3_d epo_d = 0;
   while ( den_u ) {
-    c3_d res_d = sscanf(den_u->nam_c, "0i%" PRIu64, &epo_d);
-    if ( (1 == res_d) && (epo_d != pre_d) && (epo_d != pos_d) ) {
-      fprintf(stderr, "chop: deleting epoch 0i%" PRIu64 "\r\n", epo_d);
-      if ( c3y != u3_disk_epoc_kill(log_u, epo_d) ) {
-        fprintf(stderr, "chop: failed to delete epoch 0i%" PRIu64 "\r\n", epo_d);
-      }
+    c3_i res_i = sscanf(den_u->nam_c, "0i%" PRIu64, &epo_d);
+    if ( (1 == res_i) && (epo_d < pre_d) && (epo_d < pos_d) ) {
     }
     den_u = den_u->nex_u;
   }
 
   // cleanup
+  u3_dire_free(ned_u);
   u3_disk_exit(log_u);
   u3m_stop();
 
@@ -2269,7 +2307,7 @@ _cw_roll(c3_i argc, c3_c* argv[])
 
   if ( fir_d == las_d ) {
     fprintf(stderr, "roll: latest epoch already empty\r\n");
-    exit(1);
+    exit(0);
   }
   else if ( c3n == u3_disk_epoc_init(log_u) ) {
     fprintf(stderr, "roll: failed to create new epoch\r\n");

From b8b52c239e4f2c3de03a119999a6c7408039f8f3 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Fri, 26 May 2023 09:43:55 -0400
Subject: [PATCH 17/59] disk: move format version to `version.h`

---
 pkg/noun/version.h | 6 ++++++
 pkg/vere/disk.c    | 6 ------
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/pkg/noun/version.h b/pkg/noun/version.h
index f64d0a398e..125abd684c 100644
--- a/pkg/noun/version.h
+++ b/pkg/noun/version.h
@@ -18,4 +18,10 @@ typedef c3_w       u3e_version;
 #define U3E_VER1   1
 #define U3E_VERLAT U3E_VER1
 
+/* DISK FORMAT
+ */
+
+#define U3D_VER1   1
+#define U3D_VERLAT U3L_VER1
+
 #endif /* ifndef U3_VERSION_H */
diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index ce6930ea89..2138ed4205 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -23,12 +23,6 @@ struct _cd_save {
   struct _u3_disk* log_u;
 };
 
-/* DISK FORMAT
- */
-
-#define U3D_VER1   1
-#define U3D_VERLAT U3L_VER1
-
 #undef VERBOSE_DISK
 #undef DISK_TRACE_JAM
 #undef DISK_TRACE_CUE

From 6582f14ad1be153d2ed636ae8077efd90fc26a03 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Tue, 30 May 2023 11:47:23 -0400
Subject: [PATCH 18/59] disk: use `PRIc3_d` instead of `PRIu64`

---
 pkg/vere/disk.c | 36 ++++++++++++++++--------------------
 pkg/vere/main.c | 10 +++++-----
 2 files changed, 21 insertions(+), 25 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 2138ed4205..27c7b14f66 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -905,7 +905,7 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
 
     //  set path to latest epoch
     c3_c epo_c[8193];
-    snprintf(epo_c, 8192, "%s/0i%" PRIu64, log_c, lat_d);
+    snprintf(epo_c, 8192, "%s/0i%" PRIc3_d, log_c, lat_d);
 
     //  initialize latest epoch's db
     if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c)) ) {
@@ -913,7 +913,7 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
       c3_free(log_u);
       return 0;
     }
-    fprintf(stderr, "disk: loaded epoch 0i%" PRIu64 "\r\n", lat_d);
+    fprintf(stderr, "disk: loaded epoch 0i%" PRIc3_d "\r\n", lat_d);
 
     //  get first/last event numbers from lmdb
     c3_d fir_d, las_d;
@@ -975,7 +975,7 @@ c3_o u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d) {
 
   //  file paths
   c3_c epo_c[8193], dat_c[8193], epv_c[8193], biv_c[8193];
-  snprintf(epo_c, sizeof(epo_c), "%s/0i%" PRIu64, log_u->com_u->pax_c, epo_d);
+  snprintf(epo_c, sizeof(epo_c), "%s/0i%" PRIc3_d, log_u->com_u->pax_c, epo_d);
   snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", epo_c);
   snprintf(epv_c, sizeof(epv_c), "%s/epoc.txt", epo_c);
   snprintf(biv_c, sizeof(biv_c), "%s/vere.txt", epo_c);
@@ -1014,23 +1014,23 @@ c3_o u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d) {
 
   //  print error messages
   if ( c3n == dir_o ) {
-    fprintf(stderr, "disk: epoch 0i%" PRIu64 " is not writable\r\n", epo_d);
+    fprintf(stderr, "disk: epoch 0i%" PRIc3_d " is not writable\r\n", epo_d);
     ret_o = c3n;
   }
   if ( c3n == dat_o ) {
-    fprintf(stderr, "disk: epoch 0i%" PRIu64 "/data.mdb is not writable\r\n", epo_d);
+    fprintf(stderr, "disk: epoch 0i%" PRIc3_d "/data.mdb is not writable\r\n", epo_d);
     ret_o = c3n;
   }
   if ( c3n == mdb_o ) {
-    fprintf(stderr, "disk: epoch 0i%" PRIu64 "/data.mdb can't be opened\r\n", epo_d);
+    fprintf(stderr, "disk: epoch 0i%" PRIc3_d "/data.mdb can't be opened\r\n", epo_d);
     ret_o = c3n;
   }
   if ( c3n == epv_o ) {
-    fprintf(stderr, "disk: epoch 0i%" PRIu64 "/epoc.txt is not writable\r\n", epo_d);
+    fprintf(stderr, "disk: epoch 0i%" PRIc3_d "/epoc.txt is not writable\r\n", epo_d);
     ret_o = c3n;
   }
   if ( c3n == biv_o ) {
-    fprintf(stderr, "disk: epoch 0i%" PRIu64 "/vere.txt is not writable\r\n", epo_d);
+    fprintf(stderr, "disk: epoch 0i%" PRIc3_d "/vere.txt is not writable\r\n", epo_d);
     ret_o = c3n;
   }
 
@@ -1063,10 +1063,10 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
 
   //  create new epoch directory if it doesn't exist
   c3_c epo_c[8193];
-  snprintf(epo_c, sizeof(epo_c), "%s/0i%" PRIu64, log_u->com_u->pax_c, new_d);
+  snprintf(epo_c, sizeof(epo_c), "%s/0i%" PRIc3_d, log_u->com_u->pax_c, new_d);
   c3_d ret_d = c3_mkdir(epo_c, 0700);
   if ( ( ret_d < 0 ) && ( errno != EEXIST ) ) {
-    fprintf(stderr, "disk: failed to create epoch directory %" PRIu64 "\r\n", new_d);
+    fprintf(stderr, "disk: failed to create epoch directory %" PRIc3_d "\r\n", new_d);
     return c3n;
   }
 
@@ -1107,10 +1107,6 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
   if ( c3y == u3_Host.ops_u.nuu || new_d > 0 ) {
     c3_c dat_c[8193];
     snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", epo_c);
-    // if ( c3n == c3_unlink(dat_c) ) {
-    //   fprintf(stderr, "disk: failed to rm 0i%" PRIu64 "/data.mdb\r\n", new_d);
-    //   goto fail;
-    // };
     if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c)) ) {
       fprintf(stderr, "disk: failed to initialize database\r\n");
       c3_free(log_u);
@@ -1144,7 +1140,7 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
 c3_o u3_disk_epoc_kill(u3_disk* log_u, c3_d epo_d) {
   //  get epoch directory
   c3_c epo_c[8193];
-  snprintf(epo_c, sizeof(epo_c), "%s/0i%" PRIu64, log_u->com_u->pax_c, epo_d);
+  snprintf(epo_c, sizeof(epo_c), "%s/0i%" PRIc3_d, log_u->com_u->pax_c, epo_d);
 
   //  delete files in epoch directory
   u3_dire* dir_u = u3_foil_folder(epo_c);
@@ -1177,7 +1173,7 @@ c3_o u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d) {
   u3_dent* den_u = u3_foil_folder(log_u->com_u->pax_c)->dil_u;
   while ( den_u ) {
     c3_d epo_d = 0;
-    if ( 1 == sscanf(den_u->nam_c, "0i%" PRIu64, &epo_d) ) {
+    if ( 1 == sscanf(den_u->nam_c, "0i%" PRIc3_d, &epo_d) ) {
       ret_d = c3y;   //  NB: returns yes if the directory merely exists
     }
     *lat_d = c3_max(epo_d, *lat_d);  //  update the latest epoch number
@@ -1192,18 +1188,18 @@ c3_o u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d) {
 c3_o
 u3_disk_epoc_vere(u3_disk* log_u, c3_d epo_d, c3_c* ver_w) {
   c3_c ver_c[8193];
-  snprintf(ver_c, sizeof(ver_c), "%s/0i%" PRIu64 "/vere.txt", 
+  snprintf(ver_c, sizeof(ver_c), "%s/0i%" PRIc3_d "/vere.txt", 
                   log_u->com_u->pax_c, epo_d);
 
   FILE* fil_u = fopen(ver_c, "r");
   if ( NULL == fil_u ) {
-    fprintf(stderr, "disk: failed to open vere.txt in epoch 0i%" PRIu64 
+    fprintf(stderr, "disk: failed to open vere.txt in epoch 0i%" PRIc3_d 
                     "\r\n", epo_d);
     return c3n;
   }
 
   if ( 1 != fscanf(fil_u, "%s", ver_w) ) {
-    fprintf(stderr, "disk: failed to read vere.txt in epoch 0i%" PRIu64 
+    fprintf(stderr, "disk: failed to read vere.txt in epoch 0i%" PRIc3_d 
                     "\r\n", epo_d);
     return c3n;
   }
@@ -1283,7 +1279,7 @@ c3_o u3_disk_migrate(u3_disk* log_u)
     if ( u3_Host.eve_d != las_d ) {
       fprintf(stderr, "disk: migrate: error: snapshot is out of date, please "
                       "start/shutdown your pier gracefully first\r\n");
-      fprintf(stderr, "disk: migrate: eve_d (%" PRIu64 ") != las_d (%" PRIu64 ")\r\n",
+      fprintf(stderr, "disk: migrate: eve_d (%" PRIc3_d ") != las_d (%" PRIc3_d ")\r\n",
                       u3_Host.eve_d, las_d);
       return c3n;
     }
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index 7f9d360e9a..71bd981cbb 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -2140,7 +2140,7 @@ _cw_chop(c3_i argc, c3_c* argv[])
   if ( log_u->dun_d != u3A->eve_d ) {
     fprintf(stderr, "chop: error: snapshot is out of date, please "
                     "start/shutdown your pier gracefully first\r\n");
-    fprintf(stderr, "chop: eve_d: %" PRIu64 ", dun_d: %" PRIu64 "\r\n", u3A->eve_d, log_u->dun_d);
+    fprintf(stderr, "chop: eve_d: %" PRIc3_d ", dun_d: %" PRIc3_d "\r\n", u3A->eve_d, log_u->dun_d);
     exit(1);
   }
 
@@ -2176,7 +2176,7 @@ _cw_chop(c3_i argc, c3_c* argv[])
   len_z = 0;
   den_u = ned_u->dil_u;
   while ( den_u ) {
-    if ( 1 == sscanf(den_u->nam_c, "0i%" PRIu64, (sot_d + len_z)) ) {
+    if ( 1 == sscanf(den_u->nam_c, "0i%" PRIc3_d, (sot_d + len_z)) ) {
       len_z++;
     }
     den_u = den_u->nex_u;
@@ -2209,7 +2209,7 @@ _cw_chop(c3_i argc, c3_c* argv[])
   //  delete all but the last two epochs
   //  XX parameterize the number of epochs to chop
   for ( c3_z i_z = 2; i_z < len_z; i_z++ ) {
-    fprintf(stderr, "chop: deleting epoch 0i%" PRIu64 "\r\n", sot_d[i_z]);
+    fprintf(stderr, "chop: deleting epoch 0i%" PRIc3_d "\r\n", sot_d[i_z]);
     if ( c3y != u3_disk_epoc_kill(log_u, sot_d[i_z]) ) {
       fprintf(stderr, "chop: failed to delete epoch 0i%" PRIu64 "\r\n", sot_d[i_z]);
       exit(1);
@@ -2218,7 +2218,7 @@ _cw_chop(c3_i argc, c3_c* argv[])
 
   c3_d epo_d = 0;
   while ( den_u ) {
-    c3_i res_i = sscanf(den_u->nam_c, "0i%" PRIu64, &epo_d);
+    c3_i res_i = sscanf(den_u->nam_c, "0i%" PRIc3_d, &epo_d);
     if ( (1 == res_i) && (epo_d < pre_d) && (epo_d < pos_d) ) {
     }
     den_u = den_u->nex_u;
@@ -2293,7 +2293,7 @@ _cw_roll(c3_i argc, c3_c* argv[])
   if ( log_u->dun_d != u3A->eve_d ) {
     fprintf(stderr, "roll: error: snapshot is out of date, please "
                     "start/shutdown your pier gracefully first\r\n");
-    fprintf(stderr, "roll: eve_d: %" PRIu64 ", dun_d: %" PRIu64 "\r\n", \
+    fprintf(stderr, "roll: eve_d: %" PRIc3_d ", dun_d: %" PRIc3_d "\r\n", \
                      u3A->eve_d, log_u->dun_d);
     exit(1);
   }

From b35b413f9aaabd71924524657d91cc03dcf70144 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Wed, 31 May 2023 15:34:17 -0400
Subject: [PATCH 19/59] disk: various epoch improvements

---
 pkg/noun/manage.c |  1 +
 pkg/vere/disk.c   | 43 ++++++++++++++++++++++++++-----------------
 pkg/vere/main.c   | 11 ++---------
 3 files changed, 29 insertions(+), 26 deletions(-)

diff --git a/pkg/noun/manage.c b/pkg/noun/manage.c
index 166dcdf8df..2d467c11ce 100644
--- a/pkg/noun/manage.c
+++ b/pkg/noun/manage.c
@@ -2054,6 +2054,7 @@ extern void u3je_secp_stop(void);
 void
 u3m_stop()
 {
+  //  XX make sure to cleanup snapshot file descriptors if necessary
   u3je_secp_stop();
 }
 
diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 27c7b14f66..6f7ce2176f 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -896,8 +896,8 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
     }
 
     //  get binary version from latest epoch
-    c3_c ver_w[8193];
-    if ( c3n == u3_disk_epoc_vere(log_u, lat_d, ver_w) ) {
+    c3_c ver_c[8193];
+    if ( c3n == u3_disk_epoc_vere(log_u, lat_d, ver_c) ) {
       fprintf(stderr, "disk: failed to load epoch version\r\n");
       c3_free(log_u);
       return 0;
@@ -936,7 +936,7 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
 
     //  if binary version of latest epoch is not the same as the
     //  running binary, then we need to create a new epoch
-    if ( 0 != strcmp(ver_w, URBIT_VERSION) ) {
+    if ( 0 != strcmp(ver_c, URBIT_VERSION) ) {
       if ( c3n == u3_disk_epoc_init(log_u) ) {
         fprintf(stderr, "disk: failed to initialize epoch\r\n");
         c3_free(log_u);
@@ -1161,6 +1161,9 @@ c3_o u3_disk_epoc_kill(u3_disk* log_u, c3_d epo_d) {
     return c3n;
   }
 
+  //  cleanup
+  u3_dire_free(dir_u);
+
   //  success
   return c3y;
 }
@@ -1168,25 +1171,31 @@ c3_o u3_disk_epoc_kill(u3_disk* log_u, c3_d epo_d) {
 /* u3_disk_epoc_last: get latest epoch number.
 */
 c3_o u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d) {
-  c3_o ret_d = c3n;  //  return no if no epoch directories exist
+  c3_o ret_o = c3n;  //  return no if no epoch directories exist
   *lat_d = 0;        //  initialize lat_d to 0
-  u3_dent* den_u = u3_foil_folder(log_u->com_u->pax_c)->dil_u;
+  u3_dire* die_u = u3_foil_folder(log_u->com_u->pax_c);
+  u3_dent* den_u = die_u->dil_u;
   while ( den_u ) {
     c3_d epo_d = 0;
     if ( 1 == sscanf(den_u->nam_c, "0i%" PRIc3_d, &epo_d) ) {
-      ret_d = c3y;   //  NB: returns yes if the directory merely exists
+      ret_o = c3y;   //  NB: returns yes if the directory merely exists
+      *lat_d = c3_max(epo_d, *lat_d);  //  update the latest epoch number
     }
-    *lat_d = c3_max(epo_d, *lat_d);  //  update the latest epoch number
     den_u = den_u->nex_u;
   }
+  
+  u3_dire_free(die_u);
 
-  return ret_d;
+  return ret_o;
 }
 
 /* u3_disk_epoc_vere: get binary version from epoch.
 */
 c3_o
 u3_disk_epoc_vere(u3_disk* log_u, c3_d epo_d, c3_c* ver_w) {
+  //  XX  check implementation (similar code is in king.c, _king_get_pace)
+  //      should probably write a generic function to read a string from 
+  //      a file into malloc'd memory, with an optional whitespace trimming
   c3_c ver_c[8193];
   snprintf(ver_c, sizeof(ver_c), "%s/0i%" PRIc3_d "/vere.txt", 
                   log_u->com_u->pax_c, epo_d);
@@ -1210,12 +1219,12 @@ u3_disk_epoc_vere(u3_disk* log_u, c3_d epo_d, c3_c* ver_w) {
  */
 c3_o u3_disk_migrate(u3_disk* log_u)
 {
-  /*  migration steps:
+  /*  migration steps (* indicates breakpoint set):
    *  0. detect whether we need to migrate or not
-   *     a. if it's a fresh boot via u3_Host.ops_u.nuu -> skip migration
-   *     b. if data.mdb is readable in log directory -> execute migration
-   *        if not -> skip migration
-   *  1. initialize epoch 0i0 (first call to u3_disk_epoc_init())
+   *     a. if it's a fresh boot via u3_Host.ops_u.nuu -> skip migration (returns yes)
+   *     b. if data.mdb is readable in log directory -> execute migration (returns yes or no)
+   *        if not -> skip migration (returns yes)
+   *  1. initialize epoch 0i0 (first call to u3_disk_epoc_init()) *
    *     a. creates epoch directory
    *     b. creates epoch version file
    *     c. creates binary version file
@@ -1224,10 +1233,10 @@ c3_o u3_disk_migrate(u3_disk* log_u)
    *     f. writes metadata to new database
    *     g. loads new epoch directory and sets it in log_u
    *  2. create hard links to data.mdb and lock.mdb in 0i0/
-   *  3. rollover to new epoch (second call to u3_disk_epoc_init())
+   *  3. rollover to new epoch (second call to u3_disk_epoc_init()) *
    *     a. same as 1a-g but also copies current snapshot between c/d steps
-   *  4. delete backup snapshot (c3_unlink() and c3_rmdir() calls)
-   *  5. delete old data.mdb and lock.mdb files (c3_unlink() calls)
+   *  4. delete backup snapshot (c3_unlink() and c3_rmdir() calls) *
+   *  5. delete old data.mdb and lock.mdb files (c3_unlink() calls) *
    */
 
   //  check if data.mdb is readable in log directory
@@ -1252,7 +1261,7 @@ c3_o u3_disk_migrate(u3_disk* log_u)
 
     //  initialize first epoch "0i0"
     if ( c3n == u3_disk_epoc_init(log_u) ) {
-      fprintf(stderr, "disk: migrate: failed to initialize first epoch\r\n");
+      fprintf(stderr, "disk: failed to initialize first epoch\r\n");
       return c3n;
     }
   } else if ( c3y == dat_o ) {
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index 71bd981cbb..79a1a0fe92 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -2216,14 +2216,6 @@ _cw_chop(c3_i argc, c3_c* argv[])
     }
   }
 
-  c3_d epo_d = 0;
-  while ( den_u ) {
-    c3_i res_i = sscanf(den_u->nam_c, "0i%" PRIc3_d, &epo_d);
-    if ( (1 == res_i) && (epo_d < pre_d) && (epo_d < pos_d) ) {
-    }
-    den_u = den_u->nex_u;
-  }
-
   // cleanup
   u3_dire_free(ned_u);
   u3_disk_exit(log_u);
@@ -2769,7 +2761,8 @@ main(c3_i   argc,
     //  validate whether we can execute disk migration
     if ( u3_Host.ops_u.nuu == c3n ) {
       u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
-      //  XX u3e_close()
+      u3m_stop();
+      //  XX  unmap loom, else parts of the snapshot could be left in memory
     }
 
     //  starting u3m configures OpenSSL memory functions, so we must do it

From ba1a3af7f1f7ee122e7cdb56382975730b701b32 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Wed, 31 May 2023 18:01:43 -0400
Subject: [PATCH 20/59] disk: function style fixes

---
 pkg/vere/disk.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 6f7ce2176f..9a5f8dfacc 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -958,7 +958,8 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
 
 /* u3_disk_epoc_good: check for valid epoch.
 */
-c3_o u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d) {
+c3_o
+u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d) {
   /*  a "valid" epoch is currently defined as a writable folder in 
    *  the <pier>/.urb/log directory named with a @ui and contains:
    *  - a writable data.mdb file which can be opened by lmdb (add db checking?)
@@ -1039,7 +1040,8 @@ c3_o u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d) {
  
 /* u3_disk_epoc_init: create new epoch.
 */
-c3_o u3_disk_epoc_init(u3_disk* log_u) {
+c3_o
+u3_disk_epoc_init(u3_disk* log_u) {
   //  set new epoch number
   c3_d lat_d, new_d;
   c3_o eps_o = u3_disk_epoc_last(log_u, &lat_d);
@@ -1137,7 +1139,8 @@ c3_o u3_disk_epoc_init(u3_disk* log_u) {
 
 /* u3_disk_epoc_kill: delete an epoch.
 */
-c3_o u3_disk_epoc_kill(u3_disk* log_u, c3_d epo_d) {
+c3_o
+u3_disk_epoc_kill(u3_disk* log_u, c3_d epo_d) {
   //  get epoch directory
   c3_c epo_c[8193];
   snprintf(epo_c, sizeof(epo_c), "%s/0i%" PRIc3_d, log_u->com_u->pax_c, epo_d);
@@ -1170,7 +1173,8 @@ c3_o u3_disk_epoc_kill(u3_disk* log_u, c3_d epo_d) {
 
 /* u3_disk_epoc_last: get latest epoch number.
 */
-c3_o u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d) {
+c3_o
+u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d) {
   c3_o ret_o = c3n;  //  return no if no epoch directories exist
   *lat_d = 0;        //  initialize lat_d to 0
   u3_dire* die_u = u3_foil_folder(log_u->com_u->pax_c);
@@ -1358,4 +1362,4 @@ c3_o u3_disk_migrate(u3_disk* log_u)
   }
 
   return c3y;
-}
\ No newline at end of file
+}

From 9b45ad20ec0d0f94437b6afe2ec8eec387459687 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Thu, 1 Jun 2023 09:14:16 -0400
Subject: [PATCH 21/59] u3: remove snapshot backup in `u3e_save` and unused
 `u3m_backup`

---
 pkg/noun/events.c |  4 ----
 pkg/noun/manage.c | 10 ----------
 2 files changed, 14 deletions(-)

diff --git a/pkg/noun/events.c b/pkg/noun/events.c
index 5e16182307..287ac92d82 100644
--- a/pkg/noun/events.c
+++ b/pkg/noun/events.c
@@ -1379,10 +1379,6 @@ u3e_save(u3_post low_p, u3_post hig_p)
                         pgs_w, ptr_v, strerror(errno));
     }
   }
-
-  c3_c bhk_c[8193];
-  snprintf(bhk_c, sizeof(bhk_c), "%s/.urb/bhk", u3P.dir_c);
-  u3e_backup(bhk_c, c3n);
 }
 
 /* u3e_live(): start the checkpointing system.
diff --git a/pkg/noun/manage.c b/pkg/noun/manage.c
index 2d467c11ce..ee93ed5568 100644
--- a/pkg/noun/manage.c
+++ b/pkg/noun/manage.c
@@ -1748,16 +1748,6 @@ _cm_limits(void)
 # endif
 }
 
-/* u3m_backup(): copy snapshot to .urb/bhk (if it doesn't exist yet).
-*/
-c3_o
-u3m_backup(c3_o ovw_o)
-{
-  c3_c bhk_c[2048];
-  snprintf(bhk_c, 2048, "%s/.urb/bhk", u3P.dir_c);
-  return u3e_backup(bhk_c, ovw_o);
-}
-
 /* u3m_fault(): handle a memory event with libsigsegv protocol.
 */
 c3_i

From fb762b2fb24846a8a4e50d852c4ca18149df3c9c Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Thu, 1 Jun 2023 09:40:45 -0400
Subject: [PATCH 22/59] disk: add event check to `u3_disk_epoc_good`

---
 pkg/vere/disk.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 9a5f8dfacc..70661b6634 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -983,6 +983,7 @@ u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d) {
 
   c3_o dir_o = c3n;  //  directory is writable
   c3_o dat_o = c3n;  //  data.mdb is writable
+  c3_o len_o = c3n;  //  data.mdb has events
   c3_o mdb_o = c3n;  //  data.mdb can be opened
   c3_o epv_o = c3n;  //  epoc.txt is writable
   c3_o biv_o = c3n;  //  vere.txt is writable
@@ -1000,6 +1001,17 @@ u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d) {
     if ( 0 != (env_u = u3_lmdb_init(epo_c)) ) {
       mdb_o = c3y;
     }
+
+    //  check if there are any events in the database
+    c3_d low_d, hig_d;
+    if ( 0 != u3_lmdb_gulf(env_u, &low_d, &hig_d) ) {
+      fprintf(stderr, "disk: failed to get first/last event numbers\r\n");
+      ret_o = c3n;
+    } else {
+      if ( low_d == hig_d == 0 ) {
+        len_o = c3y;  //  yes, data.mdb has events
+      }
+    }
     u3_lmdb_exit(env_u);
   }
 
@@ -1026,6 +1038,10 @@ u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d) {
     fprintf(stderr, "disk: epoch 0i%" PRIc3_d "/data.mdb can't be opened\r\n", epo_d);
     ret_o = c3n;
   }
+  if ( c3n == len_o ) {
+    fprintf(stderr, "disk: epoch 0i%" PRIc3_d "/data.mdb has no events\r\n", epo_d);
+    ret_o = c3n;
+  }
   if ( c3n == epv_o ) {
     fprintf(stderr, "disk: epoch 0i%" PRIc3_d "/epoc.txt is not writable\r\n", epo_d);
     ret_o = c3n;

From 83b57376bcfc4a54d9959ebc5e027b68d268daf1 Mon Sep 17 00:00:00 2001
From: barter-simsum <barter-simsum@protonmail.com>
Date: Thu, 1 Jun 2023 10:18:14 -0400
Subject: [PATCH 23/59] cleanup whitespace

---
 pkg/noun/events.c |  2 +-
 pkg/vere/disk.c   | 22 +++++++++++-----------
 pkg/vere/main.c   |  4 ++--
 pkg/vere/vere.h   | 10 +++++-----
 4 files changed, 19 insertions(+), 19 deletions(-)

diff --git a/pkg/noun/events.c b/pkg/noun/events.c
index 287ac92d82..89013cecd9 100644
--- a/pkg/noun/events.c
+++ b/pkg/noun/events.c
@@ -1220,7 +1220,7 @@ u3e_backup(c3_c* pax_c, c3_o ovw_o)
 {
   u3e_image nop_u = { .nam_c = "north", .pgs_w = 0 };
   u3e_image sop_u = { .nam_c = "south", .pgs_w = 0 };
-  
+
   c3_i mod_i = O_RDWR | O_CREAT;
 
   if ( !pax_c ) {
diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 70661b6634..6e18eebef9 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -929,8 +929,8 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
       } else {                        //  not first epoch
         log_u->dun_d = lat_d;         //  set dun_d to last event in prev epoch
       }
-    } else {                          //  not fresh epoch            
-      log_u->dun_d = las_d;           //  set dun_d to last event in lmdb        
+    } else {                          //  not fresh epoch
+      log_u->dun_d = las_d;           //  set dun_d to last event in lmdb
     }
     log_u->sen_d = log_u->dun_d;
 
@@ -960,7 +960,7 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
 */
 c3_o
 u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d) {
-  /*  a "valid" epoch is currently defined as a writable folder in 
+  /*  a "valid" epoch is currently defined as a writable folder in
    *  the <pier>/.urb/log directory named with a @ui and contains:
    *  - a writable data.mdb file which can be opened by lmdb (add db checking?)
    *  - a writable epoc.txt file
@@ -968,7 +968,7 @@ u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d) {
    *
    *  XX: should we check if the correct event sequence exists in the data.mdb file?
    *
-   *  note that this does not check if the epoch even contains snapshot files, 
+   *  note that this does not check if the epoch even contains snapshot files,
    *  let alone whether they're valid or not
    */
 
@@ -1053,7 +1053,7 @@ u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d) {
 
   return ret_o;
 }
- 
+
 /* u3_disk_epoc_init: create new epoch.
 */
 c3_o
@@ -1203,7 +1203,7 @@ u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d) {
     }
     den_u = den_u->nex_u;
   }
-  
+
   u3_dire_free(die_u);
 
   return ret_o;
@@ -1214,21 +1214,21 @@ u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d) {
 c3_o
 u3_disk_epoc_vere(u3_disk* log_u, c3_d epo_d, c3_c* ver_w) {
   //  XX  check implementation (similar code is in king.c, _king_get_pace)
-  //      should probably write a generic function to read a string from 
+  //      should probably write a generic function to read a string from
   //      a file into malloc'd memory, with an optional whitespace trimming
   c3_c ver_c[8193];
-  snprintf(ver_c, sizeof(ver_c), "%s/0i%" PRIc3_d "/vere.txt", 
+  snprintf(ver_c, sizeof(ver_c), "%s/0i%" PRIc3_d "/vere.txt",
                   log_u->com_u->pax_c, epo_d);
 
   FILE* fil_u = fopen(ver_c, "r");
   if ( NULL == fil_u ) {
-    fprintf(stderr, "disk: failed to open vere.txt in epoch 0i%" PRIc3_d 
+    fprintf(stderr, "disk: failed to open vere.txt in epoch 0i%" PRIc3_d
                     "\r\n", epo_d);
     return c3n;
   }
 
   if ( 1 != fscanf(fil_u, "%s", ver_w) ) {
-    fprintf(stderr, "disk: failed to read vere.txt in epoch 0i%" PRIc3_d 
+    fprintf(stderr, "disk: failed to read vere.txt in epoch 0i%" PRIc3_d
                     "\r\n", epo_d);
     return c3n;
   }
@@ -1286,7 +1286,7 @@ c3_o u3_disk_migrate(u3_disk* log_u)
     }
   } else if ( c3y == dat_o ) {
     //  migrate existing pier which has either:
-    //  - not started the migration, or 
+    //  - not started the migration, or
     //  - crashed before completing the migration
 
     //  initialize pre-migrated lmdb
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index 79a1a0fe92..cacb37d285 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -317,7 +317,7 @@ _main_getopt(c3_i argc, c3_c** argv)
           return c3n;
         } else {
           u3_Host.ops_u.sap_w = arg_w * 60;
-          if ( 0 == u3_Host.ops_u.sap_w) 
+          if ( 0 == u3_Host.ops_u.sap_w)
             return c3n;
         }
         break;
@@ -2757,7 +2757,7 @@ main(c3_i   argc,
       }
     }
 
-    //  we need the current snapshot's latest event number to 
+    //  we need the current snapshot's latest event number to
     //  validate whether we can execute disk migration
     if ( u3_Host.ops_u.nuu == c3n ) {
       u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
diff --git a/pkg/vere/vere.h b/pkg/vere/vere.h
index fbb28c1ef7..f3e47a98b6 100644
--- a/pkg/vere/vere.h
+++ b/pkg/vere/vere.h
@@ -988,27 +988,27 @@
        */
        c3_o
        u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d);
-      
+
       /* u3_disk_epoc_init(): create new epoch.
        */
        c3_o
        u3_disk_epoc_init(u3_disk* log_u);
-      
+
       /* u3_disk_epoc_kill(): delete an epoch.
        */
        c3_o
        u3_disk_epoc_kill(u3_disk* log_u, c3_d epo_d);
-      
+
       /* u3_disk_epoc_last(): get latest epoch number.
        */
        c3_o
        u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d);
-      
+
       /* u3_disk_epoc_vere(): get binary version from epoch.
        */
        c3_o
        u3_disk_epoc_vere(u3_disk* log_u, c3_d epo_d, c3_c* ver_w);
-      
+
       /* u3_disk_migrate(): migrates disk format.
        */
         c3_o

From 6dc1bd6e0a8537e2ba6c3eedf97ee68b85139fc4 Mon Sep 17 00:00:00 2001
From: barter-simsum <barter-simsum@protonmail.com>
Date: Thu, 1 Jun 2023 11:21:45 -0400
Subject: [PATCH 24/59] data.mdb -> lock.mdb

---
 pkg/vere/disk.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 6e18eebef9..56383a76b2 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -1270,7 +1270,7 @@ c3_o u3_disk_migrate(u3_disk* log_u)
   //  check if lock.mdb is readable in log directory
   c3_o lok_o = c3n;
   c3_c lok_c[8193];
-  snprintf(lok_c, sizeof(dat_c), "%s/data.mdb", log_u->com_u->pax_c);
+  snprintf(lok_c, sizeof(dat_c), "%s/lock.mdb", log_u->com_u->pax_c);
   if ( 0 == access(dat_c, R_OK) ) {
     lok_o = c3y;
   }

From a0d239bb0d85f8084d96bb10092c6910b8ae873d Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Thu, 1 Jun 2023 14:09:17 -0400
Subject: [PATCH 25/59] disk: add event check in `u3_disk_epoc_good`

---
 pkg/vere/disk.c | 178 +++++++++++++++++++++++++-----------------------
 1 file changed, 93 insertions(+), 85 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 56383a76b2..f0807bbedf 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -959,15 +959,14 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
 /* u3_disk_epoc_good: check for valid epoch.
 */
 c3_o
-u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d) {
+u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d)
+{
   /*  a "valid" epoch is currently defined as a writable folder in
    *  the <pier>/.urb/log directory named with a @ui and contains:
-   *  - a writable data.mdb file which can be opened by lmdb (add db checking?)
+   *  - a writable data.mdb file which can be opened by lmdb and contains 1+ events
    *  - a writable epoc.txt file
    *  - a writable vere.txt file
    *
-   *  XX: should we check if the correct event sequence exists in the data.mdb file?
-   *
    *  note that this does not check if the epoch even contains snapshot files,
    *  let alone whether they're valid or not
    */
@@ -1057,7 +1056,8 @@ u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d) {
 /* u3_disk_epoc_init: create new epoch.
 */
 c3_o
-u3_disk_epoc_init(u3_disk* log_u) {
+u3_disk_epoc_init(u3_disk* log_u)
+{
   //  set new epoch number
   c3_d lat_d, new_d;
   c3_o eps_o = u3_disk_epoc_last(log_u, &lat_d);
@@ -1156,7 +1156,8 @@ u3_disk_epoc_init(u3_disk* log_u) {
 /* u3_disk_epoc_kill: delete an epoch.
 */
 c3_o
-u3_disk_epoc_kill(u3_disk* log_u, c3_d epo_d) {
+u3_disk_epoc_kill(u3_disk* log_u, c3_d epo_d)
+{
   //  get epoch directory
   c3_c epo_c[8193];
   snprintf(epo_c, sizeof(epo_c), "%s/0i%" PRIc3_d, log_u->com_u->pax_c, epo_d);
@@ -1190,7 +1191,8 @@ u3_disk_epoc_kill(u3_disk* log_u, c3_d epo_d) {
 /* u3_disk_epoc_last: get latest epoch number.
 */
 c3_o
-u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d) {
+u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d)
+{
   c3_o ret_o = c3n;  //  return no if no epoch directories exist
   *lat_d = 0;        //  initialize lat_d to 0
   u3_dire* die_u = u3_foil_folder(log_u->com_u->pax_c);
@@ -1212,7 +1214,8 @@ u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d) {
 /* u3_disk_epoc_vere: get binary version from epoch.
 */
 c3_o
-u3_disk_epoc_vere(u3_disk* log_u, c3_d epo_d, c3_c* ver_w) {
+u3_disk_epoc_vere(u3_disk* log_u, c3_d epo_d, c3_c* ver_w)
+{
   //  XX  check implementation (similar code is in king.c, _king_get_pace)
   //      should probably write a generic function to read a string from
   //      a file into malloc'd memory, with an optional whitespace trimming
@@ -1275,8 +1278,8 @@ c3_o u3_disk_migrate(u3_disk* log_u)
     lok_o = c3y;
   }
 
+  //  if fresh boot, initialize disk v1
   if ( c3y == u3_Host.ops_u.nuu ) {
-    //  initialize disk v1 on fresh boots
     fprintf(stderr, "disk: initializing disk with v%d format\r\n", U3D_VER1);
 
     //  initialize first epoch "0i0"
@@ -1284,98 +1287,103 @@ c3_o u3_disk_migrate(u3_disk* log_u)
       fprintf(stderr, "disk: failed to initialize first epoch\r\n");
       return c3n;
     }
-  } else if ( c3y == dat_o ) {
-    //  migrate existing pier which has either:
-    //  - not started the migration, or
-    //  - crashed before completing the migration
-
-    //  initialize pre-migrated lmdb
-    {
-      if ( 0 == (log_u->mdb_u = u3_lmdb_init(log_u->com_u->pax_c)) ) {
-        fprintf(stderr, "disk: failed to initialize database\r\n");
-        return c3n;
-      }
-    }
 
-    //  get first/last event numbers from pre-migrated lmdb
-    c3_d fir_d, las_d;
-    if ( c3n == u3_lmdb_gulf(log_u->mdb_u, &fir_d, &las_d) ) {
-      fprintf(stderr, "disk: failed to get first/last event numbers\r\n");
-      return c3n;
-    }
+    return c3y;
+  }
+  
+  //  if data.mdb is not readable, skip migration
+  if ( c3n == dat_o ) return c3y;
 
-    // ensure there's a current snapshot
-    if ( u3_Host.eve_d != las_d ) {
-      fprintf(stderr, "disk: migrate: error: snapshot is out of date, please "
-                      "start/shutdown your pier gracefully first\r\n");
-      fprintf(stderr, "disk: migrate: eve_d (%" PRIc3_d ") != las_d (%" PRIc3_d ")\r\n",
-                      u3_Host.eve_d, las_d);
-      return c3n;
-    }
+  //  migrate existing pier which has either:
+  //  - not started the migration, or
+  //  - crashed before completing the migration
 
-    //  initialize first epoch "0i0"
-    if ( c3n == u3_disk_epoc_init(log_u) ) {
-      fprintf(stderr, "disk: migrate: failed to initialize first epoch\r\n");
+  //  initialize pre-migrated lmdb
+  {
+    if ( 0 == (log_u->mdb_u = u3_lmdb_init(log_u->com_u->pax_c)) ) {
+      fprintf(stderr, "disk: failed to initialize database\r\n");
       return c3n;
     }
+  }
 
-    //  create hard links to data.mdb and lock.mdb in 0i0/
-    c3_c dut_c[8193], luk_c[8193];  //  old paths
-    snprintf(dut_c, sizeof(dut_c), "%s/data.mdb", log_u->com_u->pax_c);
-    snprintf(luk_c, sizeof(luk_c), "%s/lock.mdb", log_u->com_u->pax_c);
+  //  get first/last event numbers from pre-migrated lmdb
+  c3_d fir_d, las_d;
+  if ( c3n == u3_lmdb_gulf(log_u->mdb_u, &fir_d, &las_d) ) {
+    fprintf(stderr, "disk: failed to get first/last event numbers\r\n");
+    return c3n;
+  }
 
-    c3_c epo_c[8193], dat_c[8193], lok_c[8193];  //  new paths
-    snprintf(epo_c, sizeof(epo_c), "%s/0i0", log_u->com_u->pax_c);
-    snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", epo_c);
-    snprintf(lok_c, sizeof(lok_c), "%s/lock.mdb", epo_c);
+  // ensure there's a current snapshot
+  if ( u3_Host.eve_d != las_d ) {
+    fprintf(stderr, "disk: migrate: error: snapshot is out of date, please "
+                    "start/shutdown your pier gracefully first\r\n");
+    fprintf(stderr, "disk: migrate: eve_d (%" PRIc3_d ") != las_d (%" PRIc3_d ")\r\n",
+                    u3_Host.eve_d, las_d);
+    return c3n;
+  }
 
-    if ( 0 < c3_link(dut_c, dat_c) ) {
-      fprintf(stderr, "disk: migrate: failed to create data.mdb hard link\r\n");
-      fprintf(stderr, "errno %d: %s\r\n", errno, strerror(errno));
-      return c3n;
-    }
-    if ( c3y == lok_o ) {  //  only link lock.mdb if it exists
-      if ( 0 < c3_link(luk_c, lok_c) ) {
-        fprintf(stderr, "disk: migrate: failed to create lock.mdb hard link\r\n");
-        c3_rename(dat_c, dut_c);
-        return c3n;
-      }
-    }
+  //  initialize first epoch "0i0"
+  if ( c3n == u3_disk_epoc_init(log_u) ) {
+    fprintf(stderr, "disk: migrate: failed to initialize first epoch\r\n");
+    return c3n;
+  }
 
-    //  rollover to new epoch
-    // log_u->mdb_u = old_u;
-    if ( c3n == u3_disk_epoc_init(log_u) ) {
-      fprintf(stderr, "roll: error: failed to initialize new epoch\r\n");
+  //  create hard links to data.mdb and lock.mdb in 0i0/
+  c3_c dut_c[8193], luk_c[8193];  //  old paths
+  snprintf(dut_c, sizeof(dut_c), "%s/data.mdb", log_u->com_u->pax_c);
+  snprintf(luk_c, sizeof(luk_c), "%s/lock.mdb", log_u->com_u->pax_c);
+
+  c3_c epo_c[8193];
+  snprintf(epo_c, sizeof(epo_c), "%s/0i0", log_u->com_u->pax_c);
+  snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", epo_c);
+  snprintf(lok_c, sizeof(lok_c), "%s/lock.mdb", epo_c);
+
+  if ( 0 < c3_link(dut_c, dat_c) ) {
+    fprintf(stderr, "disk: migrate: failed to create data.mdb hard link\r\n");
+    fprintf(stderr, "errno %d: %s\r\n", errno, strerror(errno));
+    return c3n;
+  }
+  if ( c3y == lok_o ) {  //  only link lock.mdb if it exists
+    if ( 0 < c3_link(luk_c, lok_c) ) {
+      fprintf(stderr, "disk: migrate: failed to create lock.mdb hard link\r\n");
+      c3_rename(dat_c, dut_c);
       return c3n;
     }
+  }
 
-    //  delete backup snapshot
-    c3_c bhk_c[8193], nop_c[8193], sop_c[8193];
-    snprintf(bhk_c, sizeof(bhk_c), "%s/.urb/bhk", u3_Host.dir_c);
-    snprintf(nop_c, sizeof(nop_c), "%s/north.bin", bhk_c);
-    snprintf(sop_c, sizeof(sop_c), "%s/south.bin", bhk_c);
-    if ( c3n == c3_unlink(nop_c) ) {
-      fprintf(stderr, "disk: migrate: failed to delete bhk/north.bin\r\n");
-    } else if ( c3n == c3_unlink(sop_c) ) {
-      fprintf(stderr, "disk: migrate: failed to delete bhk/south.bin\r\n");
-    } else {
-      if ( c3n == c3_rmdir(bhk_c) ) {
-        fprintf(stderr, "disk: migrate: failed to delete bhk/\r\n");
-      }
-    }
+  //  rollover to new epoch
+  // log_u->mdb_u = old_u;
+  if ( c3n == u3_disk_epoc_init(log_u) ) {
+    fprintf(stderr, "roll: error: failed to initialize new epoch\r\n");
+    return c3n;
+  }
 
-    //  delete old lock.mdb and data.mdb files
-    if ( 0 != c3_unlink(luk_c) ) {
-      fprintf(stderr, "disk: migrate: failed to unlink lock.mdb\r\n");
-    }
-    if ( 0 != c3_unlink(dut_c) ) {
-      fprintf(stderr, "disk: migrate: failed to unlink data.mdb\r\n");
-      return c3n;  //  migration succeeds only if we can unlink data.mdb
+  //  delete backup snapshot
+  c3_c bhk_c[8193], nop_c[8193], sop_c[8193];
+  snprintf(bhk_c, sizeof(bhk_c), "%s/.urb/bhk", u3_Host.dir_c);
+  snprintf(nop_c, sizeof(nop_c), "%s/north.bin", bhk_c);
+  snprintf(sop_c, sizeof(sop_c), "%s/south.bin", bhk_c);
+  if ( c3n == c3_unlink(nop_c) ) {
+    fprintf(stderr, "disk: migrate: failed to delete bhk/north.bin\r\n");
+  } else if ( c3n == c3_unlink(sop_c) ) {
+    fprintf(stderr, "disk: migrate: failed to delete bhk/south.bin\r\n");
+  } else {
+    if ( c3n == c3_rmdir(bhk_c) ) {
+      fprintf(stderr, "disk: migrate: failed to delete bhk/\r\n");
     }
+  }
 
-    //  success
-    fprintf(stderr, "disk: migrated disk to v%d format\r\n", U3D_VER1);
+  //  delete old lock.mdb and data.mdb files
+  if ( 0 != c3_unlink(luk_c) ) {
+    fprintf(stderr, "disk: migrate: failed to unlink lock.mdb\r\n");
   }
+  if ( 0 != c3_unlink(dut_c) ) {
+    fprintf(stderr, "disk: migrate: failed to unlink data.mdb\r\n");
+    return c3n;  //  migration succeeds only if we can unlink data.mdb
+  }
+
+  //  success
+  fprintf(stderr, "disk: migrated disk to v%d format\r\n", U3D_VER1);
 
   return c3y;
 }

From d6fcb9875f800c8583241736a972f41ef580ef56 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Fri, 2 Jun 2023 10:36:17 -0400
Subject: [PATCH 26/59] disk: fix duplicate code

---
 pkg/vere/disk.c | 32 ++++++++++++++------------------
 1 file changed, 14 insertions(+), 18 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index f0807bbedf..5046a3a1c6 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -1263,19 +1263,19 @@ c3_o u3_disk_migrate(u3_disk* log_u)
    */
 
   //  check if data.mdb is readable in log directory
-  c3_o dat_o = c3n;
-  c3_c dat_c[8193];
-  snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", log_u->com_u->pax_c);
-  if ( 0 == access(dat_c, R_OK) ) {
-    dat_o = c3y;
+  c3_o dut_o = c3n;
+  c3_c dut_c[8193];
+  snprintf(dut_c, sizeof(dut_c), "%s/data.mdb", log_u->com_u->pax_c);
+  if ( 0 == access(dut_c, R_OK) ) {
+    dut_o = c3y;
   }
 
   //  check if lock.mdb is readable in log directory
-  c3_o lok_o = c3n;
-  c3_c lok_c[8193];
-  snprintf(lok_c, sizeof(dat_c), "%s/lock.mdb", log_u->com_u->pax_c);
-  if ( 0 == access(dat_c, R_OK) ) {
-    lok_o = c3y;
+  c3_o luk_o = c3n;
+  c3_c luk_c[8193];
+  snprintf(luk_c, sizeof(luk_c), "%s/lock.mdb", log_u->com_u->pax_c);
+  if ( 0 == access(luk_c, R_OK) ) {
+    luk_o = c3y;
   }
 
   //  if fresh boot, initialize disk v1
@@ -1291,8 +1291,8 @@ c3_o u3_disk_migrate(u3_disk* log_u)
     return c3y;
   }
   
-  //  if data.mdb is not readable, skip migration
-  if ( c3n == dat_o ) return c3y;
+  //  if .urb/log/data.mdb is not readable, skip migration
+  if ( c3n == dut_o ) return c3y;
 
   //  migrate existing pier which has either:
   //  - not started the migration, or
@@ -1329,11 +1329,7 @@ c3_o u3_disk_migrate(u3_disk* log_u)
   }
 
   //  create hard links to data.mdb and lock.mdb in 0i0/
-  c3_c dut_c[8193], luk_c[8193];  //  old paths
-  snprintf(dut_c, sizeof(dut_c), "%s/data.mdb", log_u->com_u->pax_c);
-  snprintf(luk_c, sizeof(luk_c), "%s/lock.mdb", log_u->com_u->pax_c);
-
-  c3_c epo_c[8193];
+  c3_c epo_c[8193], dat_c[8193], lok_c[8193];
   snprintf(epo_c, sizeof(epo_c), "%s/0i0", log_u->com_u->pax_c);
   snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", epo_c);
   snprintf(lok_c, sizeof(lok_c), "%s/lock.mdb", epo_c);
@@ -1343,7 +1339,7 @@ c3_o u3_disk_migrate(u3_disk* log_u)
     fprintf(stderr, "errno %d: %s\r\n", errno, strerror(errno));
     return c3n;
   }
-  if ( c3y == lok_o ) {  //  only link lock.mdb if it exists
+  if ( c3y == luk_o ) {  //  only link lock.mdb if it exists
     if ( 0 < c3_link(luk_c, lok_c) ) {
       fprintf(stderr, "disk: migrate: failed to create lock.mdb hard link\r\n");
       c3_rename(dat_c, dut_c);

From 1886844f4abbd4aac66cc91ec80c661d256c1817 Mon Sep 17 00:00:00 2001
From: barter-simsum <barter-simsum@protonmail.com>
Date: Fri, 2 Jun 2023 10:48:33 -0400
Subject: [PATCH 27/59] remove redundant dut_o in u3_disk_migrate

---
 pkg/vere/disk.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 5046a3a1c6..81eb3a600d 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -1263,11 +1263,12 @@ c3_o u3_disk_migrate(u3_disk* log_u)
    */
 
   //  check if data.mdb is readable in log directory
-  c3_o dut_o = c3n;
   c3_c dut_c[8193];
   snprintf(dut_c, sizeof(dut_c), "%s/data.mdb", log_u->com_u->pax_c);
-  if ( 0 == access(dut_c, R_OK) ) {
-    dut_o = c3y;
+  if ( !_(u3_Host.ops_u.nuu)
+       && 0 != access(dut_c, R_OK) ) {
+    // if .urb/log/data.mdb is not readable, skip migration
+    return c3y;
   }
 
   //  check if lock.mdb is readable in log directory
@@ -1290,9 +1291,6 @@ c3_o u3_disk_migrate(u3_disk* log_u)
 
     return c3y;
   }
-  
-  //  if .urb/log/data.mdb is not readable, skip migration
-  if ( c3n == dut_o ) return c3y;
 
   //  migrate existing pier which has either:
   //  - not started the migration, or

From d59b0af5497add0f3c36a72d4deb86d138e0453a Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Mon, 5 Jun 2023 11:05:40 -0400
Subject: [PATCH 28/59] disk: make `u3_disk_epoc_vere` safer

---
 pkg/vere/disk.c | 33 +++++++++++++++++++++++----------
 1 file changed, 23 insertions(+), 10 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 81eb3a600d..eb1b25792c 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -1216,25 +1216,38 @@ u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d)
 c3_o
 u3_disk_epoc_vere(u3_disk* log_u, c3_d epo_d, c3_c* ver_w)
 {
-  //  XX  check implementation (similar code is in king.c, _king_get_pace)
-  //      should probably write a generic function to read a string from
-  //      a file into malloc'd memory, with an optional whitespace trimming
-  c3_c ver_c[8193];
-  snprintf(ver_c, sizeof(ver_c), "%s/0i%" PRIc3_d "/vere.txt",
-                  log_u->com_u->pax_c, epo_d);
-
-  FILE* fil_u = fopen(ver_c, "r");
-  if ( NULL == fil_u ) {
+  struct stat buf_u;
+  c3_c* ver_c;
+  c3_w red_w, len_w;
+  c3_i ret_i, fid_i;
+  ret_i = asprintf(&ver_c, "%s/0i%" PRIc3_d "/vere.txt",
+           log_u->com_u->pax_c, epo_d);
+  u3_assert( ret_i > 0 );
+
+  fid_i = c3_open(ver_c, O_RDONLY, 0644);
+
+  if ( (fid_i < 0) || (fstat(fid_i, &buf_u) < 0) ) {
     fprintf(stderr, "disk: failed to open vere.txt in epoch 0i%" PRIc3_d
                     "\r\n", epo_d);
     return c3n;
   }
 
-  if ( 1 != fscanf(fil_u, "%s", ver_w) ) {
+  len_w = buf_u.st_size;
+  red_w = read(fid_i, ver_w, len_w);
+  close(fid_i);
+
+  if ( len_w != red_w ) {
     fprintf(stderr, "disk: failed to read vere.txt in epoch 0i%" PRIc3_d
                     "\r\n", epo_d);
     return c3n;
   }
+
+  //  trim trailing whitespace
+  ver_w[len_w] = 0;
+  while ( len_w-- && isspace(ver_w[len_w]) ) {
+    ver_w[len_w] = 0;
+  }
+
   return c3y;
 }
 

From c50906de6e69cf11a94ed9df245b51371783858b Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Mon, 5 Jun 2023 11:29:17 -0400
Subject: [PATCH 29/59] disk: cleanup `#include` directives

---
 pkg/noun/noun.h | 1 -
 pkg/noun/urth.c | 1 -
 pkg/vere/disk.c | 1 +
 3 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/pkg/noun/noun.h b/pkg/noun/noun.h
index b493307288..965330a519 100644
--- a/pkg/noun/noun.h
+++ b/pkg/noun/noun.h
@@ -15,7 +15,6 @@
 #include "types.h"
 #include "vortex.h"
 #include "zave.h"
-#include "events.h"
 #include "imprison.h"
 #include "log.h"
 #include "nock.h"
diff --git a/pkg/noun/urth.c b/pkg/noun/urth.c
index a4a52e2522..05ca4db33c 100644
--- a/pkg/noun/urth.c
+++ b/pkg/noun/urth.c
@@ -8,7 +8,6 @@
 #include <sys/stat.h>
 
 #include "allocate.h"
-#include "events.h"
 #include "hashtable.h"
 #include "imprison.h"
 #include "jets.h"
diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index eb1b25792c..23be66ba4b 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -1,6 +1,7 @@
 /// @file
 
 #include "noun.h"
+#include "events.h"
 #include "vere.h"
 #include "version.h"
 #include "db/lmdb.h"

From 611856662e9624bbe3f0a4f3466af209d44c87fc Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Mon, 5 Jun 2023 11:34:53 -0400
Subject: [PATCH 30/59] lmdb: restore `siz_i` parameter in `u3_lmdb_init`

---
 pkg/vere/db/lmdb.c | 11 +----------
 pkg/vere/db/lmdb.h |  2 +-
 pkg/vere/disk.c    | 36 ++++++++++++++++++++++++++++++++----
 3 files changed, 34 insertions(+), 15 deletions(-)

diff --git a/pkg/vere/db/lmdb.c b/pkg/vere/db/lmdb.c
index d9fed4c0de..8f8864b77f 100644
--- a/pkg/vere/db/lmdb.c
+++ b/pkg/vere/db/lmdb.c
@@ -38,20 +38,11 @@ intmax_t mdb_get_filesize(mdb_filehandle_t han_u);
 /* u3_lmdb_init(): open lmdb at [pax_c], mmap up to [siz_i].
 */
 MDB_env*
-u3_lmdb_init(const c3_c* pax_c)
+u3_lmdb_init(const c3_c* pax_c, size_t siz_i)
 {
   MDB_env* env_u;
   c3_w     ret_w;
 
-  //  calculate db size based on architecture
-  const size_t siz_i =
-  // 500 GiB is as large as musl on aarch64 wants to allow
-  #if (defined(U3_CPU_aarch64) && defined(U3_OS_linux))
-    0x7d00000000;
-  #else
-    0x10000000000;
-  #endif
-
   if ( (ret_w = mdb_env_create(&env_u)) ) {
     mdb_logerror(stderr, ret_w, "lmdb: init fail");
     return 0;
diff --git a/pkg/vere/db/lmdb.h b/pkg/vere/db/lmdb.h
index 7aa996001f..aab4e6c9f7 100644
--- a/pkg/vere/db/lmdb.h
+++ b/pkg/vere/db/lmdb.h
@@ -22,7 +22,7 @@
     /* u3_lmdb_init(): open lmdb at [pax_c], mmap up to [siz_i].
     */
       MDB_env*
-      u3_lmdb_init(const c3_c* pax_c);
+      u3_lmdb_init(const c3_c* pax_c, size_t siz_i);
 
     /* u3_lmdb_exit(): close lmdb.
     */
diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 23be66ba4b..d34c5b852a 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -909,7 +909,14 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
     snprintf(epo_c, 8192, "%s/0i%" PRIc3_d, log_c, lat_d);
 
     //  initialize latest epoch's db
-    if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c)) ) {
+    const size_t siz_i =
+    // 500 GiB is as large as musl on aarch64 wants to allow
+    #if (defined(U3_CPU_aarch64) && defined(U3_OS_linux))
+      0x7d00000000;
+    #else
+      0x10000000000;
+    #endif
+    if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c, siz_i)) ) {
       fprintf(stderr, "disk: failed to initialize database\r\n");
       c3_free(log_u);
       return 0;
@@ -998,7 +1005,14 @@ u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d)
     dat_o = c3y;
     //  check if we can open data.mdb
     MDB_env* env_u;
-    if ( 0 != (env_u = u3_lmdb_init(epo_c)) ) {
+    const size_t siz_i =
+    // 500 GiB is as large as musl on aarch64 wants to allow
+    #if (defined(U3_CPU_aarch64) && defined(U3_OS_linux))
+      0x7d00000000;
+    #else
+      0x10000000000;
+    #endif
+    if ( 0 != (env_u = u3_lmdb_init(epo_c, siz_i)) ) {
       mdb_o = c3y;
     }
 
@@ -1126,7 +1140,14 @@ u3_disk_epoc_init(u3_disk* log_u)
   if ( c3y == u3_Host.ops_u.nuu || new_d > 0 ) {
     c3_c dat_c[8193];
     snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", epo_c);
-    if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c)) ) {
+    const size_t siz_i =
+    // 500 GiB is as large as musl on aarch64 wants to allow
+    #if (defined(U3_CPU_aarch64) && defined(U3_OS_linux))
+      0x7d00000000;
+    #else
+      0x10000000000;
+    #endif
+    if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c, siz_i)) ) {
       fprintf(stderr, "disk: failed to initialize database\r\n");
       c3_free(log_u);
       goto fail;
@@ -1312,7 +1333,14 @@ c3_o u3_disk_migrate(u3_disk* log_u)
 
   //  initialize pre-migrated lmdb
   {
-    if ( 0 == (log_u->mdb_u = u3_lmdb_init(log_u->com_u->pax_c)) ) {
+    const size_t siz_i =
+    // 500 GiB is as large as musl on aarch64 wants to allow
+    #if (defined(U3_CPU_aarch64) && defined(U3_OS_linux))
+      0x7d00000000;
+    #else
+      0x10000000000;
+    #endif
+    if ( 0 == (log_u->mdb_u = u3_lmdb_init(log_u->com_u->pax_c, siz_i)) ) {
       fprintf(stderr, "disk: failed to initialize database\r\n");
       return c3n;
     }

From d5aff59e419ff5f4674eb7c6471dc5d2d2924218 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Tue, 6 Jun 2023 14:29:37 -0400
Subject: [PATCH 31/59] disk: define `siz_i` as a global, static constant

---
 pkg/vere/disk.c | 39 +++++++++------------------------------
 1 file changed, 9 insertions(+), 30 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index d34c5b852a..7ad6420a6c 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -24,6 +24,15 @@ struct _cd_save {
   struct _u3_disk* log_u;
 };
 
+// for u3_lmdb_init() calls
+static const size_t siz_i =
+#if (defined(U3_CPU_aarch64) && defined(U3_OS_linux))
+  // 500 GiB is as large as musl on aarch64 wants to allow
+  0x7d00000000;
+#else
+  0x10000000000;
+#endif
+
 #undef VERBOSE_DISK
 #undef DISK_TRACE_JAM
 #undef DISK_TRACE_CUE
@@ -909,13 +918,6 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
     snprintf(epo_c, 8192, "%s/0i%" PRIc3_d, log_c, lat_d);
 
     //  initialize latest epoch's db
-    const size_t siz_i =
-    // 500 GiB is as large as musl on aarch64 wants to allow
-    #if (defined(U3_CPU_aarch64) && defined(U3_OS_linux))
-      0x7d00000000;
-    #else
-      0x10000000000;
-    #endif
     if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c, siz_i)) ) {
       fprintf(stderr, "disk: failed to initialize database\r\n");
       c3_free(log_u);
@@ -1005,13 +1007,6 @@ u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d)
     dat_o = c3y;
     //  check if we can open data.mdb
     MDB_env* env_u;
-    const size_t siz_i =
-    // 500 GiB is as large as musl on aarch64 wants to allow
-    #if (defined(U3_CPU_aarch64) && defined(U3_OS_linux))
-      0x7d00000000;
-    #else
-      0x10000000000;
-    #endif
     if ( 0 != (env_u = u3_lmdb_init(epo_c, siz_i)) ) {
       mdb_o = c3y;
     }
@@ -1140,13 +1135,6 @@ u3_disk_epoc_init(u3_disk* log_u)
   if ( c3y == u3_Host.ops_u.nuu || new_d > 0 ) {
     c3_c dat_c[8193];
     snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", epo_c);
-    const size_t siz_i =
-    // 500 GiB is as large as musl on aarch64 wants to allow
-    #if (defined(U3_CPU_aarch64) && defined(U3_OS_linux))
-      0x7d00000000;
-    #else
-      0x10000000000;
-    #endif
     if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c, siz_i)) ) {
       fprintf(stderr, "disk: failed to initialize database\r\n");
       c3_free(log_u);
@@ -1333,13 +1321,6 @@ c3_o u3_disk_migrate(u3_disk* log_u)
 
   //  initialize pre-migrated lmdb
   {
-    const size_t siz_i =
-    // 500 GiB is as large as musl on aarch64 wants to allow
-    #if (defined(U3_CPU_aarch64) && defined(U3_OS_linux))
-      0x7d00000000;
-    #else
-      0x10000000000;
-    #endif
     if ( 0 == (log_u->mdb_u = u3_lmdb_init(log_u->com_u->pax_c, siz_i)) ) {
       fprintf(stderr, "disk: failed to initialize database\r\n");
       return c3n;
@@ -1382,13 +1363,11 @@ c3_o u3_disk_migrate(u3_disk* log_u)
   if ( c3y == luk_o ) {  //  only link lock.mdb if it exists
     if ( 0 < c3_link(luk_c, lok_c) ) {
       fprintf(stderr, "disk: migrate: failed to create lock.mdb hard link\r\n");
-      c3_rename(dat_c, dut_c);
       return c3n;
     }
   }
 
   //  rollover to new epoch
-  // log_u->mdb_u = old_u;
   if ( c3n == u3_disk_epoc_init(log_u) ) {
     fprintf(stderr, "roll: error: failed to initialize new epoch\r\n");
     return c3n;

From cc2557cd8e11018d38d319afebdf6c8df0fed8f9 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Fri, 9 Jun 2023 16:09:34 -0400
Subject: [PATCH 32/59] disk: pare down `u3_disk_epoc_good`

---
 pkg/vere/disk.c | 104 +++++++++++++-----------------------------------
 1 file changed, 27 insertions(+), 77 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 7ad6420a6c..9459fecedc 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -5,6 +5,7 @@
 #include "vere.h"
 #include "version.h"
 #include "db/lmdb.h"
+#include <types.h>
 
 struct _cd_read {
   uv_timer_t       tim_u;
@@ -905,14 +906,6 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
       return 0;
     }
 
-    //  get binary version from latest epoch
-    c3_c ver_c[8193];
-    if ( c3n == u3_disk_epoc_vere(log_u, lat_d, ver_c) ) {
-      fprintf(stderr, "disk: failed to load epoch version\r\n");
-      c3_free(log_u);
-      return 0;
-    }
-
     //  set path to latest epoch
     c3_c epo_c[8193];
     snprintf(epo_c, 8192, "%s/0i%" PRIc3_d, log_c, lat_d);
@@ -946,6 +939,12 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
 
     //  if binary version of latest epoch is not the same as the
     //  running binary, then we need to create a new epoch
+    c3_c ver_c[8193];
+    if ( c3n == u3_disk_epoc_vere(log_u, lat_d, ver_c) ) {
+      fprintf(stderr, "disk: failed to load epoch version\r\n");
+      c3_free(log_u);
+      return 0;
+    }
     if ( 0 != strcmp(ver_c, URBIT_VERSION) ) {
       if ( c3n == u3_disk_epoc_init(log_u) ) {
         fprintf(stderr, "disk: failed to initialize epoch\r\n");
@@ -971,14 +970,10 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
 c3_o
 u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d)
 {
-  /*  a "valid" epoch is currently defined as a writable folder in
-   *  the <pier>/.urb/log directory named with a @ui and contains:
-   *  - a writable data.mdb file which can be opened by lmdb and contains 1+ events
-   *  - a writable epoc.txt file
-   *  - a writable vere.txt file
-   *
-   *  note that this does not check if the epoch even contains snapshot files,
-   *  let alone whether they're valid or not
+  /*  this function is used to check whether or not we can overwrite 
+   *  an epoch. if the epoch contains an openable data.mdb file that 
+   *  also contains events, then we consider it good and *cannot* 
+   *  overwrite it. if not, then we consider it bad and *can* overwrite it.
    */
 
   c3_o ret_o = c3y;  //  return code
@@ -987,78 +982,33 @@ u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d)
   c3_c epo_c[8193], dat_c[8193], epv_c[8193], biv_c[8193];
   snprintf(epo_c, sizeof(epo_c), "%s/0i%" PRIc3_d, log_u->com_u->pax_c, epo_d);
   snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", epo_c);
-  snprintf(epv_c, sizeof(epv_c), "%s/epoc.txt", epo_c);
-  snprintf(biv_c, sizeof(biv_c), "%s/vere.txt", epo_c);
 
-  c3_o dir_o = c3n;  //  directory is writable
-  c3_o dat_o = c3n;  //  data.mdb is writable
-  c3_o len_o = c3n;  //  data.mdb has events
-  c3_o mdb_o = c3n;  //  data.mdb can be opened
-  c3_o epv_o = c3n;  //  epoc.txt is writable
-  c3_o biv_o = c3n;  //  vere.txt is writable
-
-  //  check if dir is writable
-  if ( 0 == access(epo_c, W_OK) ) {
-    dir_o = c3y;
-  }
-
-  //  check if data.mdb is writable
-  if ( 0 == access(dat_c, W_OK) ) {
-    dat_o = c3y;
-    //  check if we can open data.mdb
-    MDB_env* env_u;
-    if ( 0 != (env_u = u3_lmdb_init(epo_c, siz_i)) ) {
-      mdb_o = c3y;
-    }
+  c3_o len_o = c3n;  //  data.mdb is empty
 
-    //  check if there are any events in the database
-    c3_d low_d, hig_d;
-    if ( 0 != u3_lmdb_gulf(env_u, &low_d, &hig_d) ) {
-      fprintf(stderr, "disk: failed to get first/last event numbers\r\n");
-      ret_o = c3n;
-    } else {
-      if ( low_d == hig_d == 0 ) {
-        len_o = c3y;  //  yes, data.mdb has events
-      }
-    }
-    u3_lmdb_exit(env_u);
-  }
-
-  //  check if epoc.txt is writable
-  if ( 0 == access(epv_c, W_OK) ) {
-    epv_o = c3y;
+  //  check if we can open data.mdb
+  MDB_env* env_u;
+  if ( 0 == (env_u = u3_lmdb_init(epo_c, siz_i)) ) {
+    fprintf(stderr, "disk: failed to open epoch 0i%" PRIc3_d "/data.mdb\r\n", epo_d);
+    exit(1);
   }
 
-  //  check if vere.txt is writable
-  if ( 0 == access(biv_c, W_OK) ) {
-    biv_o = c3y;
+  //  check if there are any events in the database
+  c3_d low_d, hig_d;
+  if ( 0 != u3_lmdb_gulf(env_u, &low_d, &hig_d) ) {
+    fprintf(stderr, "disk: failed to get first/last event numbers\r\n");
+    ret_o = c3n;
+  } else {
+    if ( low_d == hig_d == 0 ) {
+      len_o = c3y;  //  yes, data.mdb is empty
+    }
   }
+  u3_lmdb_exit(env_u);
 
   //  print error messages
-  if ( c3n == dir_o ) {
-    fprintf(stderr, "disk: epoch 0i%" PRIc3_d " is not writable\r\n", epo_d);
-    ret_o = c3n;
-  }
-  if ( c3n == dat_o ) {
-    fprintf(stderr, "disk: epoch 0i%" PRIc3_d "/data.mdb is not writable\r\n", epo_d);
-    ret_o = c3n;
-  }
-  if ( c3n == mdb_o ) {
-    fprintf(stderr, "disk: epoch 0i%" PRIc3_d "/data.mdb can't be opened\r\n", epo_d);
-    ret_o = c3n;
-  }
   if ( c3n == len_o ) {
     fprintf(stderr, "disk: epoch 0i%" PRIc3_d "/data.mdb has no events\r\n", epo_d);
     ret_o = c3n;
   }
-  if ( c3n == epv_o ) {
-    fprintf(stderr, "disk: epoch 0i%" PRIc3_d "/epoc.txt is not writable\r\n", epo_d);
-    ret_o = c3n;
-  }
-  if ( c3n == biv_o ) {
-    fprintf(stderr, "disk: epoch 0i%" PRIc3_d "/vere.txt is not writable\r\n", epo_d);
-    ret_o = c3n;
-  }
 
   return ret_o;
 }

From 8703f11fd12df5883e5b24f05b7dee6e2103aa96 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Mon, 12 Jun 2023 10:23:28 -0400
Subject: [PATCH 33/59] disk: remove `u3_disk_epoc_good` and add argument to
 `u3_disk_epoc_init`

---
 pkg/vere/disk.c | 114 ++++++++++--------------------------------------
 pkg/vere/main.c |   4 +-
 pkg/vere/vere.h |   7 +--
 3 files changed, 27 insertions(+), 98 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 9459fecedc..41da47653d 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -946,7 +946,7 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
       return 0;
     }
     if ( 0 != strcmp(ver_c, URBIT_VERSION) ) {
-      if ( c3n == u3_disk_epoc_init(log_u) ) {
+      if ( c3n == u3_disk_epoc_init(log_u, log_u->dun_d) ) {
         fprintf(stderr, "disk: failed to initialize epoch\r\n");
         c3_free(log_u);
         return 0;
@@ -965,86 +965,21 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
   return log_u;
 }
 
-/* u3_disk_epoc_good: check for valid epoch.
-*/
-c3_o
-u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d)
-{
-  /*  this function is used to check whether or not we can overwrite 
-   *  an epoch. if the epoch contains an openable data.mdb file that 
-   *  also contains events, then we consider it good and *cannot* 
-   *  overwrite it. if not, then we consider it bad and *can* overwrite it.
-   */
-
-  c3_o ret_o = c3y;  //  return code
-
-  //  file paths
-  c3_c epo_c[8193], dat_c[8193], epv_c[8193], biv_c[8193];
-  snprintf(epo_c, sizeof(epo_c), "%s/0i%" PRIc3_d, log_u->com_u->pax_c, epo_d);
-  snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", epo_c);
-
-  c3_o len_o = c3n;  //  data.mdb is empty
-
-  //  check if we can open data.mdb
-  MDB_env* env_u;
-  if ( 0 == (env_u = u3_lmdb_init(epo_c, siz_i)) ) {
-    fprintf(stderr, "disk: failed to open epoch 0i%" PRIc3_d "/data.mdb\r\n", epo_d);
-    exit(1);
-  }
-
-  //  check if there are any events in the database
-  c3_d low_d, hig_d;
-  if ( 0 != u3_lmdb_gulf(env_u, &low_d, &hig_d) ) {
-    fprintf(stderr, "disk: failed to get first/last event numbers\r\n");
-    ret_o = c3n;
-  } else {
-    if ( low_d == hig_d == 0 ) {
-      len_o = c3y;  //  yes, data.mdb is empty
-    }
-  }
-  u3_lmdb_exit(env_u);
-
-  //  print error messages
-  if ( c3n == len_o ) {
-    fprintf(stderr, "disk: epoch 0i%" PRIc3_d "/data.mdb has no events\r\n", epo_d);
-    ret_o = c3n;
-  }
-
-  return ret_o;
-}
-
 /* u3_disk_epoc_init: create new epoch.
 */
 c3_o
-u3_disk_epoc_init(u3_disk* log_u)
+u3_disk_epoc_init(u3_disk* log_u, c3_d epo_d)
 {
-  //  set new epoch number
-  c3_d lat_d, new_d;
+  //  check if any epoch directories exist
+  c3_d lat_d;
   c3_o eps_o = u3_disk_epoc_last(log_u, &lat_d);
-  if ( c3n == eps_o ) {
-    //  no epochs yet, so create first one 0i0
-    new_d = 0;
-  } else if ( c3n == u3_disk_epoc_good(log_u, lat_d) ) {
-    //  last epoch is invalid, so overwrite it
-    new_d = lat_d;
-  } else {
-    //  last epoch is valid, so create next one
-
-    //  get first/last event numbers
-    c3_d fir_d, las_d;
-    if ( c3n == u3_lmdb_gulf(log_u->mdb_u, &fir_d, &las_d) ) {
-      fprintf(stderr, "disk: failed to get first/last event numbers\r\n");
-      return c3n;
-    }
-    new_d = las_d;  //  create next epoch
-  }
 
   //  create new epoch directory if it doesn't exist
   c3_c epo_c[8193];
-  snprintf(epo_c, sizeof(epo_c), "%s/0i%" PRIc3_d, log_u->com_u->pax_c, new_d);
+  snprintf(epo_c, sizeof(epo_c), "%s/0i%" PRIc3_d, log_u->com_u->pax_c, epo_d);
   c3_d ret_d = c3_mkdir(epo_c, 0700);
   if ( ( ret_d < 0 ) && ( errno != EEXIST ) ) {
-    fprintf(stderr, "disk: failed to create epoch directory %" PRIc3_d "\r\n", new_d);
+    fprintf(stderr, "disk: failed to create epoch directory %" PRIc3_d "\r\n", epo_d);
     return c3n;
   }
 
@@ -1063,7 +998,7 @@ u3_disk_epoc_init(u3_disk* log_u)
   fclose(biv_f);
 
   //  copy snapshot (skip if first epoch)
-  if ( new_d > 0 ) {
+  if ( epo_d > 0 ) {
     if ( c3n == u3e_backup(epo_c, c3y) ) {
       fprintf(stderr, "disk: failed to copy snapshot to new epoch\r\n");
       goto fail;
@@ -1074,7 +1009,7 @@ u3_disk_epoc_init(u3_disk* log_u)
   c3_d     who_d[2];
   c3_o     fak_o;
   c3_w     lif_w;
-  if ( c3y == eps_o ) {
+  if ( c3y == eps_o ) {  //  skip if no epochs yet
     if ( c3y != u3_disk_read_meta(log_u->mdb_u, who_d, &fak_o, &lif_w) ) {
       fprintf(stderr, "disk: failed to read metadata\r\n");
       goto fail;
@@ -1082,7 +1017,7 @@ u3_disk_epoc_init(u3_disk* log_u)
   }
 
   //  initialize db of new epoch
-  if ( c3y == u3_Host.ops_u.nuu || new_d > 0 ) {
+  if ( c3y == u3_Host.ops_u.nuu || epo_d > 0 ) {
     c3_c dat_c[8193];
     snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", epo_c);
     if ( 0 == (log_u->mdb_u = u3_lmdb_init(epo_c, siz_i)) ) {
@@ -1153,7 +1088,7 @@ u3_disk_epoc_kill(u3_disk* log_u, c3_d epo_d)
 c3_o
 u3_disk_epoc_last(u3_disk* log_u, c3_d* lat_d)
 {
-  c3_o ret_o = c3n;  //  return no if no epoch directories exist
+  c3_o ret_o = c3n;  //  return c3n if no epoch directories exist
   *lat_d = 0;        //  initialize lat_d to 0
   u3_dire* die_u = u3_foil_folder(log_u->com_u->pax_c);
   u3_dent* den_u = die_u->dil_u;
@@ -1257,7 +1192,7 @@ c3_o u3_disk_migrate(u3_disk* log_u)
     fprintf(stderr, "disk: initializing disk with v%d format\r\n", U3D_VER1);
 
     //  initialize first epoch "0i0"
-    if ( c3n == u3_disk_epoc_init(log_u) ) {
+    if ( c3n == u3_disk_epoc_init(log_u, 0) ) {
       fprintf(stderr, "disk: failed to initialize first epoch\r\n");
       return c3n;
     }
@@ -1286,16 +1221,16 @@ c3_o u3_disk_migrate(u3_disk* log_u)
 
   // ensure there's a current snapshot
   if ( u3_Host.eve_d != las_d ) {
-    fprintf(stderr, "disk: migrate: error: snapshot is out of date, please "
+    fprintf(stderr, "disk: snapshot is out of date, please "
                     "start/shutdown your pier gracefully first\r\n");
-    fprintf(stderr, "disk: migrate: eve_d (%" PRIc3_d ") != las_d (%" PRIc3_d ")\r\n",
+    fprintf(stderr, "disk: eve_d (%" PRIc3_d ") != las_d (%" PRIc3_d ")\r\n",
                     u3_Host.eve_d, las_d);
     return c3n;
   }
 
   //  initialize first epoch "0i0"
-  if ( c3n == u3_disk_epoc_init(log_u) ) {
-    fprintf(stderr, "disk: migrate: failed to initialize first epoch\r\n");
+  if ( c3n == u3_disk_epoc_init(log_u, 0) ) {
+    fprintf(stderr, "disk: failed to initialize first epoch\r\n");
     return c3n;
   }
 
@@ -1306,20 +1241,19 @@ c3_o u3_disk_migrate(u3_disk* log_u)
   snprintf(lok_c, sizeof(lok_c), "%s/lock.mdb", epo_c);
 
   if ( 0 < c3_link(dut_c, dat_c) ) {
-    fprintf(stderr, "disk: migrate: failed to create data.mdb hard link\r\n");
-    fprintf(stderr, "errno %d: %s\r\n", errno, strerror(errno));
+    fprintf(stderr, "disk: failed to create data.mdb hard link\r\n");
     return c3n;
   }
   if ( c3y == luk_o ) {  //  only link lock.mdb if it exists
     if ( 0 < c3_link(luk_c, lok_c) ) {
-      fprintf(stderr, "disk: migrate: failed to create lock.mdb hard link\r\n");
+      fprintf(stderr, "disk: failed to create lock.mdb hard link\r\n");
       return c3n;
     }
   }
 
   //  rollover to new epoch
-  if ( c3n == u3_disk_epoc_init(log_u) ) {
-    fprintf(stderr, "roll: error: failed to initialize new epoch\r\n");
+  if ( c3n == u3_disk_epoc_init(log_u, las_d) ) {
+    fprintf(stderr, "disk: failed to initialize new epoch\r\n");
     return c3n;
   }
 
@@ -1329,21 +1263,21 @@ c3_o u3_disk_migrate(u3_disk* log_u)
   snprintf(nop_c, sizeof(nop_c), "%s/north.bin", bhk_c);
   snprintf(sop_c, sizeof(sop_c), "%s/south.bin", bhk_c);
   if ( c3n == c3_unlink(nop_c) ) {
-    fprintf(stderr, "disk: migrate: failed to delete bhk/north.bin\r\n");
+    fprintf(stderr, "disk: failed to delete bhk/north.bin\r\n");
   } else if ( c3n == c3_unlink(sop_c) ) {
-    fprintf(stderr, "disk: migrate: failed to delete bhk/south.bin\r\n");
+    fprintf(stderr, "disk: failed to delete bhk/south.bin\r\n");
   } else {
     if ( c3n == c3_rmdir(bhk_c) ) {
-      fprintf(stderr, "disk: migrate: failed to delete bhk/\r\n");
+      fprintf(stderr, "disk: failed to delete bhk/\r\n");
     }
   }
 
   //  delete old lock.mdb and data.mdb files
   if ( 0 != c3_unlink(luk_c) ) {
-    fprintf(stderr, "disk: migrate: failed to unlink lock.mdb\r\n");
+    fprintf(stderr, "disk: failed to unlink lock.mdb\r\n");
   }
   if ( 0 != c3_unlink(dut_c) ) {
-    fprintf(stderr, "disk: migrate: failed to unlink data.mdb\r\n");
+    fprintf(stderr, "disk: failed to unlink data.mdb\r\n");
     return c3n;  //  migration succeeds only if we can unlink data.mdb
   }
 
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index dc9d3819ef..d29d8fba20 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -2160,7 +2160,7 @@ _cw_chop(c3_i argc, c3_c* argv[])
   }
 
   //  create new epoch if latest isn't empty
-  if ( (fir_d != las_d) && (c3n == u3_disk_epoc_init(log_u)) ) {
+  if ( (fir_d != las_d) && (c3n == u3_disk_epoc_init(log_u, las_d)) ) {
     fprintf(stderr, "chop: failed to create new epoch\r\n");
     exit(1);
   }
@@ -2302,7 +2302,7 @@ _cw_roll(c3_i argc, c3_c* argv[])
     fprintf(stderr, "roll: latest epoch already empty\r\n");
     exit(0);
   }
-  else if ( c3n == u3_disk_epoc_init(log_u) ) {
+  else if ( c3n == u3_disk_epoc_init(log_u, las_d) ) {
     fprintf(stderr, "roll: failed to create new epoch\r\n");
     exit(1);
   }
diff --git a/pkg/vere/vere.h b/pkg/vere/vere.h
index f3e47a98b6..8d613e77b8 100644
--- a/pkg/vere/vere.h
+++ b/pkg/vere/vere.h
@@ -984,15 +984,10 @@
         void
         u3_disk_plan(u3_disk* log_u, u3_fact* tac_u);
 
-      /* u3_disk_epoc_good(): check for valid epoch.
-       */
-       c3_o
-       u3_disk_epoc_good(u3_disk* log_u, c3_d epo_d);
-
       /* u3_disk_epoc_init(): create new epoch.
        */
        c3_o
-       u3_disk_epoc_init(u3_disk* log_u);
+       u3_disk_epoc_init(u3_disk* log_u, c3_d epo_d);
 
       /* u3_disk_epoc_kill(): delete an epoch.
        */

From 52f76bc93390122aa28711f6c44b15ae560264f2 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Thu, 15 Jun 2023 14:24:05 -0400
Subject: [PATCH 34/59] wip

---
 pkg/vere/main.c | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index 0333d4d8f7..bb5b514bcc 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -2096,6 +2096,28 @@ _cw_play(c3_i argc, c3_c* argv[])
 
   u3C.wag_w |= u3o_hashless;
 
+  //  replaying events with the epoch system
+  //
+  //  there should be one process that accounts for what is actually available 
+  //  in the pier and replays the least events possible to produce a valid,
+  //  up-to-date snapshot.
+  //
+  //  this replay process also uses the chk/ folder as a place to 
+  //  "accumulate" snapshots over the course of a replay if it needs to,
+  //  overwriting the existing .bin files as it goes.
+  //
+  //  0. try loading the snapshot in chk/
+  //     a. if it is valid and up-to-date with the log
+  //        i. if we executed via `urbit play`, exit 0
+  //        ii. if we executed via `urbit`, boot into live mode
+  //     b. if it is invalid, try loading a snapshot from
+  //        i. the snapshot in the latest epoch
+  //           1. if it is valid, replay the events and save the resulting
+  //              snapshot into chk/, exit 0 or boot into live mode as before
+  //           2. if it is invalid, return an error that informs the user of 
+  //              his need for a serious surgical operation
+  //
+
   //  XX this should restore the epoch snapshot and replay that
   //
   if ( c3y == ful_o ) {

From 20a3af06cf59581e1eafa2518db0c485f3b4589e Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Tue, 20 Jun 2023 12:35:32 -0400
Subject: [PATCH 35/59] disk: set `epo_u`

---
 pkg/vere/disk.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 7371f5a3fa..b06ea35fec 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -1105,6 +1105,9 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
       }
     }
 
+    //  mark the latest epoch directory
+    log_u->epo_u = u3_foil_folder(epo_c);
+
     //  mark the log as live
     log_u->liv_o = c3y;
   }

From e5723adbe3ef4e08034eb7548ddcb5971a8bd2ba Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Wed, 21 Jun 2023 11:58:42 -0400
Subject: [PATCH 36/59] disk: `urbit play` with epochs

---
 .gitignore        |  4 +++
 pkg/noun/events.c | 92 +++++++++++++++++++++++++++++++++--------------
 pkg/noun/events.h |  5 +--
 pkg/vere/disk.c   |  6 ++--
 pkg/vere/main.c   | 19 +++++-----
 5 files changed, 85 insertions(+), 41 deletions(-)

diff --git a/.gitignore b/.gitignore
index 251a0d7f0f..da5bf8e6c8 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,3 +1,7 @@
+# IDEs.
+/.vscode
+/.idea
+
 # Bazel.
 /.user.bazelrc
 /bazel-*
diff --git a/pkg/noun/events.c b/pkg/noun/events.c
index efb506f072..6da639e476 100644
--- a/pkg/noun/events.c
+++ b/pkg/noun/events.c
@@ -312,26 +312,44 @@ _ce_image_stat(u3e_image* img_u, c3_w* pgs_w)
   }
 }
 
-/* _ce_image_open(): open or create image.
+/* _ce_image_open(): open or create image, creating directories if necessary.
 */
 static c3_o
-_ce_image_open(u3e_image* img_u)
+_ce_image_open(u3e_image* img_u, c3_c* ful_c)
 {
   c3_i mod_i = O_RDWR | O_CREAT;
-  c3_c ful_c[8193];
 
-  snprintf(ful_c, 8192, "%s", u3P.dir_c);
-  c3_mkdir(ful_c, 0700);
+  //  create directories if necessary
+  //
+  {
+    c3_c pax_c[8192];
+    c3_c* end_c = pax_c + 8192;
+    c3_c* cur_c = pax_c + snprintf(pax_c, 8192, "%s", ful_c);
 
-  snprintf(ful_c, 8192, "%s/.urb", u3P.dir_c);
-  c3_mkdir(ful_c, 0700);
+    while ( cur_c < end_c ) {
+      c3_c* nxt_c = strchr(cur_c, '/');
 
-  snprintf(ful_c, 8192, "%s/.urb/chk", u3P.dir_c);
-  c3_mkdir(ful_c, 0700);
+      if ( !nxt_c ) {
+        break;
+      }
+      else {
+        *nxt_c = 0;
+
+        if ( 0 != mkdir(pax_c, 0777) && EEXIST != errno ) {
+          fprintf(stderr, "loom: mkdir %s: %s\r\n", pax_c, strerror(errno));
+          return c3n;
+        }
 
-  snprintf(ful_c, 8192, "%s/.urb/chk/%s.bin", u3P.dir_c, img_u->nam_c);
-  if ( -1 == (img_u->fid_i = c3_open(ful_c, mod_i, 0666)) ) {
-    fprintf(stderr, "loom: c3_open %s: %s\r\n", ful_c, strerror(errno));
+        *nxt_c = '/';
+        cur_c = nxt_c + 1;
+      }
+    }
+  }
+
+  c3_c pax_c[8192];
+  snprintf(pax_c, 8192, "%s/%s.bin", ful_c, img_u->nam_c);
+  if ( -1 == (img_u->fid_i = c3_open(pax_c, mod_i, 0666)) ) {
+    fprintf(stderr, "loom: c3_open %s: %s\r\n", pax_c, strerror(errno));
     return c3n;
   }
   else if ( c3n == _ce_image_stat(img_u, &img_u->pgs_w) ) {
@@ -1231,17 +1249,24 @@ _ce_image_copy(u3e_image* fom_u, u3e_image* tou_u)
   return c3y;
 }
 
-/* u3e_backup(): copy snapshot to pax_c, overwrite optionally.
+/* u3e_backup(): copy snapshot from [pux_c] to [pax_c], 
+ * overwriting optionally. note that image files must 
+ * be named "north" and "south".
 */
 c3_o
-u3e_backup(c3_c* pax_c, c3_o ovw_o)
+u3e_backup(c3_c* pux_c, c3_c* pax_c, c3_o ovw_o)
 {
-  u3e_image nop_u = { .nam_c = "north", .pgs_w = 0 };
-  u3e_image sop_u = { .nam_c = "south", .pgs_w = 0 };
+  //  source image files from [pux_c]
+  u3e_image nux_u = { .nam_c = "north", .pgs_w = 0 };
+  u3e_image sux_u = { .nam_c = "south", .pgs_w = 0 };
+
+  //  destination image files to [pax_c]
+  u3e_image nax_u = { .nam_c = "north" };
+  u3e_image sax_u = { .nam_c = "south" };
 
   c3_i mod_i = O_RDWR | O_CREAT;
 
-  if ( !pax_c ) {
+  if ( !pux_c || !pax_c ) {
     fprintf(stderr, "loom: image backup: bad path\r\n");
     return c3n;
   }
@@ -1253,24 +1278,35 @@ u3e_backup(c3_c* pax_c, c3_o ovw_o)
     return c3n;
   }
 
+  //  open source image files
+  //
+  if ( c3n == _ce_image_open(&nux_u, pux_c) ) {
+    fprintf(stderr, "loom: couldn't open north image at %s\r\n", pux_c);
+    return c3n;
+  }
+  if ( c3n == _ce_image_open(&sux_u, pux_c) ) {
+    fprintf(stderr, "loom: couldn't open south image at %s\r\n", pux_c);
+    return c3n;
+  }
+
   c3_c nop_c[8193];
-  snprintf(nop_c, 8192, "%s/%s.bin", pax_c, nop_u.nam_c);
+  snprintf(nop_c, 8192, "%s/%s.bin", pax_c, nax_u.nam_c);
 
-  if ( -1 == (nop_u.fid_i = c3_open(nop_c, mod_i, 0666)) ) {
+  if ( -1 == (nax_u.fid_i = c3_open(nop_c, mod_i, 0666)) ) {
     fprintf(stderr, "loom: c3_open %s: %s\r\n", nop_c, strerror(errno));
     return c3n;
   }
 
   c3_c sop_c[8193];
-  snprintf(sop_c, 8192, "%s/%s.bin", pax_c, sop_u.nam_c);
+  snprintf(sop_c, 8192, "%s/%s.bin", pax_c, sax_u.nam_c);
 
-  if ( -1 == (sop_u.fid_i = c3_open(sop_c, mod_i, 0666)) ) {
+  if ( -1 == (sax_u.fid_i = c3_open(sop_c, mod_i, 0666)) ) {
     fprintf(stderr, "loom: c3_open %s: %s\r\n", sop_c, strerror(errno));
     return c3n;
   }
 
-  if (  (c3n == _ce_image_copy(&u3P.nor_u, &nop_u))
-     || (c3n == _ce_image_copy(&u3P.sou_u, &sop_u)) )
+  if (  (c3n == _ce_image_copy(&nux_u, &nax_u))
+     || (c3n == _ce_image_copy(&sux_u, &sax_u)) )
   {
     c3_unlink(nop_c);
     c3_unlink(sop_c);
@@ -1278,8 +1314,8 @@ u3e_backup(c3_c* pax_c, c3_o ovw_o)
     return c3n;
   }
 
-  close(nop_u.fid_i);
-  close(sop_u.fid_i);
+  close(nax_u.fid_i);
+  close(sax_u.fid_i);
   fprintf(stderr, "loom: image backup complete\r\n");
   return c3y;
 }
@@ -1431,8 +1467,10 @@ u3e_live(c3_o nuu_o, c3_c* dir_c)
   {
     //  Open image files.
     //
-    if ( (c3n == _ce_image_open(&u3P.nor_u)) ||
-         (c3n == _ce_image_open(&u3P.sou_u)) )
+    c3_c chk_c[8193];
+    snprintf(chk_c, 8193, "%s/.urb/chk", u3P.dir_c);
+    if ( (c3n == _ce_image_open(&u3P.nor_u, chk_c)) ||
+         (c3n == _ce_image_open(&u3P.sou_u, chk_c)) )
     {
       fprintf(stderr, "boot: image failed\r\n");
       exit(1);
diff --git a/pkg/noun/events.h b/pkg/noun/events.h
index d3a2b58a0b..462372cd28 100644
--- a/pkg/noun/events.h
+++ b/pkg/noun/events.h
@@ -73,10 +73,11 @@
 
   /** Functions.
   **/
-    /* u3e_backup(): copy the snapshot to pax_c, overwrite optional.
+    /* u3e_backup(): copy the snapshot from [pux_c] to [pax_c], 
+     * overwriting optional.
     */
       c3_o 
-      u3e_backup(c3_c* pax_c, c3_o ovw_o);
+      u3e_backup(c3_c* pux_c, c3_c* pax_c, c3_o ovw_o);
 
     /* u3e_fault(): handle a memory fault.
     */
diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index b06ea35fec..103f85a484 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -1152,9 +1152,11 @@ u3_disk_epoc_init(u3_disk* log_u, c3_d epo_d)
   fprintf(biv_f, URBIT_VERSION);
   fclose(biv_f);
 
-  //  copy snapshot (skip if first epoch)
+  //  copy snapshot files (skip if first epoch)
   if ( epo_d > 0 ) {
-    if ( c3n == u3e_backup(epo_c, c3y) ) {
+    c3_c chk_c[8193];
+    snprintf(chk_c, 8192, "%s/.urb/chk", u3_Host.dir_c);
+    if ( c3n == u3e_backup(epo_c, chk_c, c3y) ) {
       fprintf(stderr, "disk: failed to copy snapshot to new epoch\r\n");
       goto fail;
     }
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index 0f552b0252..e3da6e9291 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -2118,20 +2118,19 @@ _cw_play(c3_i argc, c3_c* argv[])
   //              his need for a serious surgical operation
   //
 
-  //  XX this should restore the epoch snapshot and replay that
-  //
   if ( c3y == ful_o ) {
+    //  copy the latest epoch's snapshot files into chk/
     u3l_log("mars: preparing for full replay");
-    u3m_init((size_t)1 << u3_Host.ops_u.lom_y);
-    u3e_live(c3n, u3_Host.dir_c);
-    u3m_foul();
-    u3m_pave(c3y);
-    u3j_boot(c3y);
-  }
-  else {
-    u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
+    c3_c chk_c[8193];
+    snprintf(chk_c, 8193, "%s/.urb/chk", u3_Host.dir_c);
+    if ( 0 != u3e_backup(log_u->epo_u->pax_c, chk_c, c3y) ) {
+      fprintf(stderr, "mars: failed to copy snapshot\r\n");
+      exit(1);
+    }
   }
 
+  u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
+
   u3C.slog_f = _cw_play_slog;
 
   {

From c9a624d2b299d290ee97cd1699f9b9e3d2e9032d Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Wed, 21 Jun 2023 12:59:03 -0400
Subject: [PATCH 37/59] events: better path naming

---
 pkg/noun/events.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/pkg/noun/events.c b/pkg/noun/events.c
index 6da639e476..f9c3904432 100644
--- a/pkg/noun/events.c
+++ b/pkg/noun/events.c
@@ -1289,27 +1289,27 @@ u3e_backup(c3_c* pux_c, c3_c* pax_c, c3_o ovw_o)
     return c3n;
   }
 
-  c3_c nop_c[8193];
-  snprintf(nop_c, 8192, "%s/%s.bin", pax_c, nax_u.nam_c);
+  c3_c nax_c[8193];
+  snprintf(nax_c, 8192, "%s/%s.bin", pax_c, nax_u.nam_c);
 
-  if ( -1 == (nax_u.fid_i = c3_open(nop_c, mod_i, 0666)) ) {
-    fprintf(stderr, "loom: c3_open %s: %s\r\n", nop_c, strerror(errno));
+  if ( -1 == (nax_u.fid_i = c3_open(nax_c, mod_i, 0666)) ) {
+    fprintf(stderr, "loom: c3_open %s: %s\r\n", nax_c, strerror(errno));
     return c3n;
   }
 
-  c3_c sop_c[8193];
-  snprintf(sop_c, 8192, "%s/%s.bin", pax_c, sax_u.nam_c);
+  c3_c sax_c[8193];
+  snprintf(sax_c, 8192, "%s/%s.bin", pax_c, sax_u.nam_c);
 
-  if ( -1 == (sax_u.fid_i = c3_open(sop_c, mod_i, 0666)) ) {
-    fprintf(stderr, "loom: c3_open %s: %s\r\n", sop_c, strerror(errno));
+  if ( -1 == (sax_u.fid_i = c3_open(sax_c, mod_i, 0666)) ) {
+    fprintf(stderr, "loom: c3_open %s: %s\r\n", sax_c, strerror(errno));
     return c3n;
   }
 
   if (  (c3n == _ce_image_copy(&nux_u, &nax_u))
      || (c3n == _ce_image_copy(&sux_u, &sax_u)) )
   {
-    c3_unlink(nop_c);
-    c3_unlink(sop_c);
+    c3_unlink(nax_c);
+    c3_unlink(sax_c);
     fprintf(stderr, "loom: image backup failed\r\n");
     return c3n;
   }

From 01ecd2e40fa4d828d4614c092b61825d442d5b41 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Thu, 22 Jun 2023 09:41:51 -0400
Subject: [PATCH 38/59] disk: fix merge error

---
 pkg/vere/disk.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 73f92ac43a..7bfb89fff8 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -1156,7 +1156,7 @@ u3_disk_epoc_init(u3_disk* log_u, c3_d epo_d)
   if ( epo_d > 0 ) {
     c3_c chk_c[8193];
     snprintf(chk_c, 8192, "%s/.urb/chk", u3_Host.dir_c);
-    if ( c3n == u3e_backup(epo_c, chk_c, c3y) ) {
+    if ( c3n == u3e_backup(chk_c, epo_c, c3y) ) {
       fprintf(stderr, "disk: failed to copy snapshot to new epoch\r\n");
       goto fail;
     }

From 4336422ae061abd7cc234d2cb6abbdfc0d14e75e Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Thu, 22 Jun 2023 14:38:23 -0400
Subject: [PATCH 39/59] events: remove unnecessary/erroneous `mkdir -p`
 implementation

---
 pkg/noun/events.c | 27 ---------------------------
 1 file changed, 27 deletions(-)

diff --git a/pkg/noun/events.c b/pkg/noun/events.c
index f3fd13e5ab..afc1d04001 100644
--- a/pkg/noun/events.c
+++ b/pkg/noun/events.c
@@ -402,33 +402,6 @@ _ce_image_open(u3e_image* img_u, c3_c* ful_c)
 {
   c3_i mod_i = O_RDWR | O_CREAT;
 
-  //  create directories if necessary
-  //
-  {
-    c3_c pax_c[8192];
-    c3_c* end_c = pax_c + 8192;
-    c3_c* cur_c = pax_c + snprintf(pax_c, 8192, "%s", ful_c);
-
-    while ( cur_c < end_c ) {
-      c3_c* nxt_c = strchr(cur_c, '/');
-
-      if ( !nxt_c ) {
-        break;
-      }
-      else {
-        *nxt_c = 0;
-
-        if ( 0 != mkdir(pax_c, 0777) && EEXIST != errno ) {
-          fprintf(stderr, "loom: mkdir %s: %s\r\n", pax_c, strerror(errno));
-          return c3n;
-        }
-
-        *nxt_c = '/';
-        cur_c = nxt_c + 1;
-      }
-    }
-  }
-
   c3_c pax_c[8192];
   snprintf(pax_c, 8192, "%s/%s.bin", ful_c, img_u->nam_c);
   if ( -1 == (img_u->fid_i = c3_open(pax_c, mod_i, 0666)) ) {

From ac3dd768a7fce2e057ab13ccf965d6dd248bc183 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Thu, 22 Jun 2023 14:41:44 -0400
Subject: [PATCH 40/59] trim trailing whitespace

---
 pkg/noun/events.c | 4 ++--
 pkg/noun/events.h | 4 ++--
 pkg/vere/main.c   | 6 +++---
 3 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/pkg/noun/events.c b/pkg/noun/events.c
index afc1d04001..3078345e16 100644
--- a/pkg/noun/events.c
+++ b/pkg/noun/events.c
@@ -1336,8 +1336,8 @@ _ce_image_copy(u3e_image* fom_u, u3e_image* tou_u)
   return c3y;
 }
 
-/* u3e_backup(): copy snapshot from [pux_c] to [pax_c], 
- * overwriting optionally. note that image files must 
+/* u3e_backup(): copy snapshot from [pux_c] to [pax_c],
+ * overwriting optionally. note that image files must
  * be named "north" and "south".
 */
 c3_o
diff --git a/pkg/noun/events.h b/pkg/noun/events.h
index 0d6a1e351b..7442c9df94 100644
--- a/pkg/noun/events.h
+++ b/pkg/noun/events.h
@@ -74,10 +74,10 @@
 
   /** Functions.
   **/
-    /* u3e_backup(): copy the snapshot from [pux_c] to [pax_c], 
+    /* u3e_backup(): copy the snapshot from [pux_c] to [pax_c],
      * overwriting optional.
     */
-      c3_o 
+      c3_o
       u3e_backup(c3_c* pux_c, c3_c* pax_c, c3_o ovw_o);
 
     /* u3e_fault(): handle a memory fault.
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index 5df8125eef..cb2b59f02c 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -2234,11 +2234,11 @@ _cw_play(c3_i argc, c3_c* argv[])
 
   //  replaying events with the epoch system
   //
-  //  there should be one process that accounts for what is actually available 
+  //  there should be one process that accounts for what is actually available
   //  in the pier and replays the least events possible to produce a valid,
   //  up-to-date snapshot.
   //
-  //  this replay process also uses the chk/ folder as a place to 
+  //  this replay process also uses the chk/ folder as a place to
   //  "accumulate" snapshots over the course of a replay if it needs to,
   //  overwriting the existing .bin files as it goes.
   //
@@ -2250,7 +2250,7 @@ _cw_play(c3_i argc, c3_c* argv[])
   //        i. the snapshot in the latest epoch
   //           1. if it is valid, replay the events and save the resulting
   //              snapshot into chk/, exit 0 or boot into live mode as before
-  //           2. if it is invalid, return an error that informs the user of 
+  //           2. if it is invalid, return an error that informs the user of
   //              his need for a serious surgical operation
   //
 

From 6c6ac0c0c4c4c37c0cc01790cdc41b402e79c5c2 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Thu, 22 Jun 2023 20:31:40 -0400
Subject: [PATCH 41/59] events: `mkdir -p` for `_ce_image_open` (needed for
 boot)

---
 pkg/noun/events.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/pkg/noun/events.c b/pkg/noun/events.c
index 3078345e16..8d7910eb26 100644
--- a/pkg/noun/events.c
+++ b/pkg/noun/events.c
@@ -402,6 +402,23 @@ _ce_image_open(u3e_image* img_u, c3_c* ful_c)
 {
   c3_i mod_i = O_RDWR | O_CREAT;
 
+  //  create directories recursively like `mkdir -p`
+  //
+  {
+    c3_c* dir_c = ful_c;
+    c3_c* end_c = ful_c + strlen(ful_c);
+
+    while ( dir_c < end_c ) {
+      if ( '/' == *dir_c ) {
+        *dir_c = 0;
+        c3_mkdir(ful_c, 0700);
+        *dir_c = '/';
+      }
+      dir_c++;
+    }
+    c3_mkdir(ful_c, 0700);
+  }
+
   c3_c pax_c[8192];
   snprintf(pax_c, 8192, "%s/%s.bin", ful_c, img_u->nam_c);
   if ( -1 == (img_u->fid_i = c3_open(pax_c, mod_i, 0666)) ) {

From 64c54543315263698d50a333b8084add9b830b1f Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Thu, 22 Jun 2023 20:45:42 -0400
Subject: [PATCH 42/59] events: prevent `0B` snapshot files from being created
 during `play`

---
 pkg/noun/events.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/pkg/noun/events.c b/pkg/noun/events.c
index 8d7910eb26..8dcf529c09 100644
--- a/pkg/noun/events.c
+++ b/pkg/noun/events.c
@@ -1382,13 +1382,21 @@ u3e_backup(c3_c* pux_c, c3_c* pax_c, c3_o ovw_o)
     return c3n;
   }
 
-  //  open source image files
+  //  open source image files if they exist
   //
-  if ( c3n == _ce_image_open(&nux_u, pux_c) ) {
+
+  //  check if file exists with access
+  //
+
+  c3_c nux_c[8193];
+  snprintf(nux_c, 8192, "%s/%s.bin", pux_c, nux_u.nam_c);
+  if ( (0 < access(nux_c, F_OK)) && (c3n == _ce_image_open(&nux_u, pux_c)) ) {
     fprintf(stderr, "loom: couldn't open north image at %s\r\n", pux_c);
     return c3n;
   }
-  if ( c3n == _ce_image_open(&sux_u, pux_c) ) {
+  c3_c sux_c[8193];
+  snprintf(sux_c, 8192, "%s/%s.bin", pux_c, sux_u.nam_c);
+  if ( (0 < access(sux_c, F_OK)) && (c3n == _ce_image_open(&sux_u, pux_c)) ) {
     fprintf(stderr, "loom: couldn't open south image at %s\r\n", pux_c);
     return c3n;
   }

From b6541b73dc6fdc7772bece5d24bf7bccca760837 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Fri, 23 Jun 2023 10:14:09 -0400
Subject: [PATCH 43/59] events: remove `mkdir -p` functionality from
 `_ce_image_open`

---
 pkg/noun/events.c | 21 ---------------------
 1 file changed, 21 deletions(-)

diff --git a/pkg/noun/events.c b/pkg/noun/events.c
index 8dcf529c09..8e53f30daa 100644
--- a/pkg/noun/events.c
+++ b/pkg/noun/events.c
@@ -402,23 +402,6 @@ _ce_image_open(u3e_image* img_u, c3_c* ful_c)
 {
   c3_i mod_i = O_RDWR | O_CREAT;
 
-  //  create directories recursively like `mkdir -p`
-  //
-  {
-    c3_c* dir_c = ful_c;
-    c3_c* end_c = ful_c + strlen(ful_c);
-
-    while ( dir_c < end_c ) {
-      if ( '/' == *dir_c ) {
-        *dir_c = 0;
-        c3_mkdir(ful_c, 0700);
-        *dir_c = '/';
-      }
-      dir_c++;
-    }
-    c3_mkdir(ful_c, 0700);
-  }
-
   c3_c pax_c[8192];
   snprintf(pax_c, 8192, "%s/%s.bin", ful_c, img_u->nam_c);
   if ( -1 == (img_u->fid_i = c3_open(pax_c, mod_i, 0666)) ) {
@@ -1384,10 +1367,6 @@ u3e_backup(c3_c* pux_c, c3_c* pax_c, c3_o ovw_o)
 
   //  open source image files if they exist
   //
-
-  //  check if file exists with access
-  //
-
   c3_c nux_c[8193];
   snprintf(nux_c, 8192, "%s/%s.bin", pux_c, nux_u.nam_c);
   if ( (0 < access(nux_c, F_OK)) && (c3n == _ce_image_open(&nux_u, pux_c)) ) {

From 299c6edb02201dbda3c366c86ceee1ca86d7d4f3 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Fri, 23 Jun 2023 10:14:40 -0400
Subject: [PATCH 44/59] disk: create `.urb/chk` on init; simplify
 `log_u->dun_d` setting

---
 pkg/vere/disk.c | 30 +++++++++++++++---------------
 1 file changed, 15 insertions(+), 15 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 7bfb89fff8..b9b5b9d1b7 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -1028,6 +1028,10 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
     strcat(dir_c, "/.urb/get");
     c3_mkdir(dir_c, 0700);
 
+    strcpy(dir_c, pax_c);
+    strcat(dir_c, "/.urb/chk");
+    c3_mkdir(dir_c, 0700);
+
     c3_free(dir_c);
   }
 
@@ -1078,14 +1082,10 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
     }
 
     //  initialize dun_d/sen_d values
-    if ( 0 == las_d ) {               //  fresh epoch (no events in lmdb yet)
-      if ( 0 == lat_d ) {             //  first epoch
-        log_u->dun_d = 0;
-      } else {                        //  not first epoch
-        log_u->dun_d = lat_d;         //  set dun_d to last event in prev epoch
-      }
-    } else {                          //  not fresh epoch
-      log_u->dun_d = las_d;           //  set dun_d to last event in lmdb
+    if ( 0 == las_d ) {      //  first epoch (no events in lmdb yet)
+      log_u->dun_d = lat_d;  //  set dun_d to last event in prev epoch
+    } else {
+      log_u->dun_d = las_d;  //  set dun_d to last event in lmdb
     }
     log_u->sen_d = log_u->dun_d;
 
@@ -1307,12 +1307,12 @@ u3_disk_epoc_vere(u3_disk* log_u, c3_d epo_d, c3_c* ver_w)
  */
 c3_o u3_disk_migrate(u3_disk* log_u)
 {
-  /*  migration steps (* indicates breakpoint set):
+  /*  migration steps:
    *  0. detect whether we need to migrate or not
-   *     a. if it's a fresh boot via u3_Host.ops_u.nuu -> skip migration (returns yes)
-   *     b. if data.mdb is readable in log directory -> execute migration (returns yes or no)
+   *     a. if it's a fresh boot via u3_Host.ops_u.nuu -> skip migration
+   *     b. if data.mdb is readable in log directory -> execute migration
    *        if not -> skip migration (returns yes)
-   *  1. initialize epoch 0i0 (first call to u3_disk_epoc_init()) *
+   *  1. initialize epoch 0i0 (first call to u3_disk_epoc_init())
    *     a. creates epoch directory
    *     b. creates epoch version file
    *     c. creates binary version file
@@ -1321,10 +1321,10 @@ c3_o u3_disk_migrate(u3_disk* log_u)
    *     f. writes metadata to new database
    *     g. loads new epoch directory and sets it in log_u
    *  2. create hard links to data.mdb and lock.mdb in 0i0/
-   *  3. rollover to new epoch (second call to u3_disk_epoc_init()) *
+   *  3. rollover to new epoch (second call to u3_disk_epoc_init())
    *     a. same as 1a-g but also copies current snapshot between c/d steps
-   *  4. delete backup snapshot (c3_unlink() and c3_rmdir() calls) *
-   *  5. delete old data.mdb and lock.mdb files (c3_unlink() calls) *
+   *  4. delete backup snapshot (c3_unlink() and c3_rmdir() calls)
+   *  5. delete old data.mdb and lock.mdb files (c3_unlink() calls)
    */
 
   //  check if data.mdb is readable in log directory

From 8a3df1dd3df59b190ebe5e70cffe40fbd278c686 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Tue, 27 Jun 2023 11:06:56 -0400
Subject: [PATCH 45/59] play: full replay when only epoch 0 exists

---
 pkg/vere/disk.c | 17 ++++++-----------
 pkg/vere/main.c | 47 ++++++++++++++++++++++-------------------------
 pkg/vere/vere.h |  2 +-
 3 files changed, 29 insertions(+), 37 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index b9b5b9d1b7..615634eb30 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -1082,15 +1082,15 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
     }
 
     //  initialize dun_d/sen_d values
-    if ( 0 == las_d ) {      //  first epoch (no events in lmdb yet)
-      log_u->dun_d = lat_d;  //  set dun_d to last event in prev epoch
-    } else {
-      log_u->dun_d = las_d;  //  set dun_d to last event in lmdb
-    }
+    log_u->dun_d = ( 0 != las_d ) ? las_d : lat_d;
     log_u->sen_d = log_u->dun_d;
 
+    //  mark the latest epoch directory
+    log_u->epo_d = lat_d;
+
     //  if binary version of latest epoch is not the same as the
     //  running binary, then we need to create a new epoch
+    //  XX move this into its own function and call it in `u3_pier_stay()`
     c3_c ver_c[8193];
     if ( c3n == u3_disk_epoc_vere(log_u, lat_d, ver_c) ) {
       fprintf(stderr, "disk: failed to load epoch version\r\n");
@@ -1105,9 +1105,6 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
       }
     }
 
-    //  mark the latest epoch directory
-    log_u->epo_u = u3_foil_folder(epo_c);
-
     //  mark the log as live
     log_u->liv_o = c3y;
   }
@@ -1193,7 +1190,7 @@ u3_disk_epoc_init(u3_disk* log_u, c3_d epo_d)
   }
 
   //  load new epoch directory and set it in log_u
-  log_u->epo_u = u3_foil_folder(epo_c);
+  log_u->epo_d = epo_d;
 
   //  success
   return c3y;
@@ -1346,8 +1343,6 @@ c3_o u3_disk_migrate(u3_disk* log_u)
 
   //  if fresh boot, initialize disk v1
   if ( c3y == u3_Host.ops_u.nuu ) {
-    fprintf(stderr, "disk: initializing disk with v%d format\r\n", U3D_VER1);
-
     //  initialize first epoch "0i0"
     if ( c3n == u3_disk_epoc_init(log_u, 0) ) {
       fprintf(stderr, "disk: failed to initialize first epoch\r\n");
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index cb2b59f02c..ee09777c29 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -2232,34 +2232,31 @@ _cw_play(c3_i argc, c3_c* argv[])
 
   u3C.wag_w |= u3o_hashless;
 
-  //  replaying events with the epoch system
-  //
-  //  there should be one process that accounts for what is actually available
-  //  in the pier and replays the least events possible to produce a valid,
-  //  up-to-date snapshot.
-  //
-  //  this replay process also uses the chk/ folder as a place to
-  //  "accumulate" snapshots over the course of a replay if it needs to,
-  //  overwriting the existing .bin files as it goes.
-  //
-  //  0. try loading the snapshot in chk/
-  //     a. if it is valid and up-to-date with the log
-  //        i. if we executed via `urbit play`, exit 0
-  //        ii. if we executed via `urbit`, boot into live mode
-  //     b. if it is invalid, try loading a snapshot from
-  //        i. the snapshot in the latest epoch
-  //           1. if it is valid, replay the events and save the resulting
-  //              snapshot into chk/, exit 0 or boot into live mode as before
-  //           2. if it is invalid, return an error that informs the user of
-  //              his need for a serious surgical operation
-  //
-
   if ( c3y == ful_o ) {
-    //  copy the latest epoch's snapshot files into chk/
     u3l_log("mars: preparing for full replay");
-    c3_c chk_c[8193];
+
+    c3_c chk_c[8193], epo_c[8193];
     snprintf(chk_c, 8193, "%s/.urb/chk", u3_Host.dir_c);
-    if ( 0 != u3e_backup(log_u->epo_u->pax_c, chk_c, c3y) ) {
+    snprintf(epo_c, 8192, "%s/0i%" PRIc3_d, log_u->com_u->pax_c, log_u->epo_d);
+
+    //  if epoch 0 is the latest, delete the snapshot files in chk/
+    if ( 0 == log_u->epo_d ) {
+      c3_c nor_c[8193], sop_c[8193];
+      snprintf(nor_c, 8193, "%s/.urb/chk/north.bin", u3_Host.dir_c);
+      snprintf(sop_c, 8193, "%s/.urb/chk/south.bin", u3_Host.dir_c);
+      if ( c3_unlink(nor_c) && (ENOENT != errno) ) {
+        fprintf(stderr, "mars: failed to unlink %s: %s\r\n",
+                        nor_c, strerror(errno));
+        exit(1);
+      }
+      if ( c3_unlink(sop_c) && (ENOENT != errno) ) {
+        fprintf(stderr, "mars: failed to unlink %s: %s\r\n",
+                        sop_c, strerror(errno));
+        exit(1);
+      }
+    }
+    else if ( 0 != u3e_backup(epo_c, chk_c, c3y) ) {
+      //  copy the latest epoch's snapshot files into chk/
       fprintf(stderr, "mars: failed to copy snapshot\r\n");
       exit(1);
     }
diff --git a/pkg/vere/vere.h b/pkg/vere/vere.h
index 4e3ea2757c..58685ac4e6 100644
--- a/pkg/vere/vere.h
+++ b/pkg/vere/vere.h
@@ -539,11 +539,11 @@
           u3_dire*         dir_u;               //  main pier directory
           u3_dire*         urb_u;               //  urbit system data
           u3_dire*         com_u;               //  log directory
-          u3_dire*         epo_u;               //  current epoch directory
           c3_o             liv_o;               //  live
           void*            mdb_u;               //  lmdb env of current epoch
           c3_d             sen_d;               //  commit requested
           c3_d             dun_d;               //  committed
+          c3_d             epo_d;               //  current epoch number
           u3_disk_cb        cb_u;               //  callbacks
           u3_read*         red_u;               //  read requests
           union {                               //  write thread/request

From fa9cdddda45008feb63fcda1bd3d397ba2deb321 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Wed, 28 Jun 2023 11:23:06 -0400
Subject: [PATCH 46/59] events/disk/pier: implement `u3_disk_vere_diff`

---
 pkg/noun/events.c | 12 ++++------
 pkg/vere/disk.c   | 49 ++++++++++++++++++++++++---------------
 pkg/vere/main.c   | 59 ++++++++++++++++++++++++++---------------------
 pkg/vere/pier.c   | 10 +++++++-
 pkg/vere/vere.h   |  5 ++++
 5 files changed, 82 insertions(+), 53 deletions(-)

diff --git a/pkg/noun/events.c b/pkg/noun/events.c
index 8e53f30daa..15db40605a 100644
--- a/pkg/noun/events.c
+++ b/pkg/noun/events.c
@@ -1344,8 +1344,8 @@ c3_o
 u3e_backup(c3_c* pux_c, c3_c* pax_c, c3_o ovw_o)
 {
   //  source image files from [pux_c]
-  u3e_image nux_u = { .nam_c = "north", .pgs_w = 0 };
-  u3e_image sux_u = { .nam_c = "south", .pgs_w = 0 };
+  u3e_image nux_u = { .nam_c = "north" };
+  u3e_image sux_u = { .nam_c = "south" };
 
   //  destination image files to [pax_c]
   u3e_image nax_u = { .nam_c = "north" };
@@ -1369,28 +1369,26 @@ u3e_backup(c3_c* pux_c, c3_c* pax_c, c3_o ovw_o)
   //
   c3_c nux_c[8193];
   snprintf(nux_c, 8192, "%s/%s.bin", pux_c, nux_u.nam_c);
-  if ( (0 < access(nux_c, F_OK)) && (c3n == _ce_image_open(&nux_u, pux_c)) ) {
+  if ( (0 != access(nux_c, F_OK)) || (c3n == _ce_image_open(&nux_u, pux_c)) ) {
     fprintf(stderr, "loom: couldn't open north image at %s\r\n", pux_c);
     return c3n;
   }
   c3_c sux_c[8193];
   snprintf(sux_c, 8192, "%s/%s.bin", pux_c, sux_u.nam_c);
-  if ( (0 < access(sux_c, F_OK)) && (c3n == _ce_image_open(&sux_u, pux_c)) ) {
+  if ( (0 != access(sux_c, F_OK)) || (c3n == _ce_image_open(&sux_u, pux_c)) ) {
     fprintf(stderr, "loom: couldn't open south image at %s\r\n", pux_c);
     return c3n;
   }
 
+  //  open destination image files
   c3_c nax_c[8193];
   snprintf(nax_c, 8192, "%s/%s.bin", pax_c, nax_u.nam_c);
-
   if ( -1 == (nax_u.fid_i = c3_open(nax_c, mod_i, 0666)) ) {
     fprintf(stderr, "loom: c3_open %s: %s\r\n", nax_c, strerror(errno));
     return c3n;
   }
-
   c3_c sax_c[8193];
   snprintf(sax_c, 8192, "%s/%s.bin", pax_c, sax_u.nam_c);
-
   if ( -1 == (sax_u.fid_i = c3_open(sax_c, mod_i, 0666)) ) {
     fprintf(stderr, "loom: c3_open %s: %s\r\n", sax_c, strerror(errno));
     return c3n;
diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 615634eb30..4d95a9516e 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -1088,23 +1088,6 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
     //  mark the latest epoch directory
     log_u->epo_d = lat_d;
 
-    //  if binary version of latest epoch is not the same as the
-    //  running binary, then we need to create a new epoch
-    //  XX move this into its own function and call it in `u3_pier_stay()`
-    c3_c ver_c[8193];
-    if ( c3n == u3_disk_epoc_vere(log_u, lat_d, ver_c) ) {
-      fprintf(stderr, "disk: failed to load epoch version\r\n");
-      c3_free(log_u);
-      return 0;
-    }
-    if ( 0 != strcmp(ver_c, URBIT_VERSION) ) {
-      if ( c3n == u3_disk_epoc_init(log_u, log_u->dun_d) ) {
-        fprintf(stderr, "disk: failed to initialize epoch\r\n");
-        c3_free(log_u);
-        return 0;
-      }
-    }
-
     //  mark the log as live
     log_u->liv_o = c3y;
   }
@@ -1416,9 +1399,11 @@ c3_o u3_disk_migrate(u3_disk* log_u)
   snprintf(sop_c, sizeof(sop_c), "%s/south.bin", bhk_c);
   if ( c3n == c3_unlink(nop_c) ) {
     fprintf(stderr, "disk: failed to delete bhk/north.bin\r\n");
-  } else if ( c3n == c3_unlink(sop_c) ) {
+  }
+  else if ( c3n == c3_unlink(sop_c) ) {
     fprintf(stderr, "disk: failed to delete bhk/south.bin\r\n");
-  } else {
+  }
+  else {
     if ( c3n == c3_rmdir(bhk_c) ) {
       fprintf(stderr, "disk: failed to delete bhk/\r\n");
     }
@@ -1438,3 +1423,29 @@ c3_o u3_disk_migrate(u3_disk* log_u)
 
   return c3y;
 }
+
+
+/* u3_disk_vere_diff(): checks if vere version mismatches latest epoch's.
+*/
+c3_o
+u3_disk_vere_diff(u3_disk* log_u)
+{
+  c3_d lat_d;
+  if ( c3n == u3_disk_epoc_last(log_u, &lat_d) ) {
+    fprintf(stderr, "disk: failed to load last epoch\r\n");
+    c3_free(log_u);
+    return 0;
+  }
+
+  c3_c ver_c[8193];
+  if ( c3n == u3_disk_epoc_vere(log_u, lat_d, ver_c) ) {
+    fprintf(stderr, "disk: failed to load epoch version\r\n");
+    c3_free(log_u);
+    return 0;
+  }
+  if ( 0 != strcmp(ver_c, URBIT_VERSION) ) {
+    return c3y;
+  }
+
+  return c3n;
+}
\ No newline at end of file
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index ee09777c29..0cc3b4afc8 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -2109,6 +2109,38 @@ _cw_play_slog(u3_noun hod)
   u3z(hod);
 }
 
+/* _cw_play_snap(): prepare snapshot for full replay.
+*/
+static void
+_cw_play_snap(u3_disk* log_u)
+{
+  c3_c chk_c[8193], epo_c[8193];
+  snprintf(chk_c, 8193, "%s/.urb/chk", u3_Host.dir_c);
+  snprintf(epo_c, 8192, "%s/0i%" PRIc3_d, log_u->com_u->pax_c, log_u->epo_d);
+
+  if ( 0 == log_u->epo_d ) {
+    //  if epoch 0 is the latest, delete the snapshot files in chk/
+    c3_c nor_c[8193], sop_c[8193];
+    snprintf(nor_c, 8193, "%s/.urb/chk/north.bin", u3_Host.dir_c);
+    snprintf(sop_c, 8193, "%s/.urb/chk/south.bin", u3_Host.dir_c);
+    if ( c3_unlink(nor_c) && (ENOENT != errno) ) {
+      fprintf(stderr, "mars: failed to unlink %s: %s\r\n",
+                      nor_c, strerror(errno));
+      exit(1);
+    }
+    if ( c3_unlink(sop_c) && (ENOENT != errno) ) {
+      fprintf(stderr, "mars: failed to unlink %s: %s\r\n",
+                      sop_c, strerror(errno));
+      exit(1);
+    }
+  }
+  else if ( 0 != u3e_backup(epo_c, chk_c, c3y) ) {
+    //  copy the latest epoch's snapshot files into chk/
+    fprintf(stderr, "mars: failed to copy snapshot\r\n");
+    exit(1);
+  }
+}
+
 /* _cw_play_exit(): exit immediately.
 */
 static void
@@ -2234,32 +2266,7 @@ _cw_play(c3_i argc, c3_c* argv[])
 
   if ( c3y == ful_o ) {
     u3l_log("mars: preparing for full replay");
-
-    c3_c chk_c[8193], epo_c[8193];
-    snprintf(chk_c, 8193, "%s/.urb/chk", u3_Host.dir_c);
-    snprintf(epo_c, 8192, "%s/0i%" PRIc3_d, log_u->com_u->pax_c, log_u->epo_d);
-
-    //  if epoch 0 is the latest, delete the snapshot files in chk/
-    if ( 0 == log_u->epo_d ) {
-      c3_c nor_c[8193], sop_c[8193];
-      snprintf(nor_c, 8193, "%s/.urb/chk/north.bin", u3_Host.dir_c);
-      snprintf(sop_c, 8193, "%s/.urb/chk/south.bin", u3_Host.dir_c);
-      if ( c3_unlink(nor_c) && (ENOENT != errno) ) {
-        fprintf(stderr, "mars: failed to unlink %s: %s\r\n",
-                        nor_c, strerror(errno));
-        exit(1);
-      }
-      if ( c3_unlink(sop_c) && (ENOENT != errno) ) {
-        fprintf(stderr, "mars: failed to unlink %s: %s\r\n",
-                        sop_c, strerror(errno));
-        exit(1);
-      }
-    }
-    else if ( 0 != u3e_backup(epo_c, chk_c, c3y) ) {
-      //  copy the latest epoch's snapshot files into chk/
-      fprintf(stderr, "mars: failed to copy snapshot\r\n");
-      exit(1);
-    }
+    _cw_play_snap(log_u);
   }
 
   u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
diff --git a/pkg/vere/pier.c b/pkg/vere/pier.c
index 263af55e1a..c2e7c47026 100644
--- a/pkg/vere/pier.c
+++ b/pkg/vere/pier.c
@@ -782,7 +782,7 @@ _pier_on_lord_wyrd_bail(void* ptr_v, u3_ovum* egg_u, u3_noun lud)
 #endif
 }
 
-/* _pier_wyrd_init(): construct %wyrd.
+/* _pier_wyrd_card(): construct %wyrd.
 */
 static u3_noun
 _pier_wyrd_card(u3_pier* pir_u)
@@ -825,6 +825,14 @@ _pier_wyrd_card(u3_pier* pir_u)
 static void
 _pier_wyrd_init(u3_pier* pir_u)
 {
+  //  create a new epoch if current version mismatches the latest epoch's
+  if ( c3y == u3_disk_vere_diff(pir_u->log_u) ) {
+    if ( c3n == u3_disk_epoc_init(pir_u->log_u, pir_u->log_u->dun_d) ) {
+      fprintf(stderr, "disk: failed to initialize epoch\r\n");
+      exit(1);
+    }
+  }
+
   u3_noun cad = _pier_wyrd_card(pir_u);
   u3_noun wir = u3nc(c3__arvo, u3_nul);
 
diff --git a/pkg/vere/vere.h b/pkg/vere/vere.h
index 58685ac4e6..c72417f86d 100644
--- a/pkg/vere/vere.h
+++ b/pkg/vere/vere.h
@@ -1025,6 +1025,11 @@
        c3_o
        u3_disk_epoc_vere(u3_disk* log_u, c3_d epo_d, c3_c* ver_w);
 
+      /* u3_disk_vere_diff(): checks if vere version mismatches latest epoch's.
+       */
+       c3_o
+       u3_disk_vere_diff(u3_disk* log_u);
+
       /* u3_disk_migrate(): migrates disk format.
        */
         c3_o

From 4d35ba25b3300b82a00b3b82b4d76d1cc9581407 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Wed, 28 Jun 2023 11:43:22 -0400
Subject: [PATCH 47/59] pier: move `.urb/chk` creation into `u3m_pier`

---
 pkg/noun/manage.c | 38 +++++++++++++++++++++++++++++++++++++-
 pkg/noun/manage.h |  5 +++++
 pkg/vere/disk.c   |  4 ----
 3 files changed, 42 insertions(+), 5 deletions(-)

diff --git a/pkg/noun/manage.c b/pkg/noun/manage.c
index 9b2450af12..2e3ce7accd 100644
--- a/pkg/noun/manage.c
+++ b/pkg/noun/manage.c
@@ -2068,6 +2068,42 @@ u3m_stop()
   u3je_secp_stop();
 }
 
+/* u3m_pier(): make a pier.
+*/
+c3_c*
+u3m_pier(c3_c* dir_c)
+{
+  c3_c ful_c[8193];
+
+  u3C.dir_c = dir_c;
+
+  snprintf(ful_c, 8192, "%s", dir_c);
+  if ( c3_mkdir(ful_c, 0700) ) {
+    if ( EEXIST != errno ) {
+      fprintf(stderr, "loom: pier create: %s\r\n", strerror(errno));
+      exit(1);
+    }
+  }
+
+  snprintf(ful_c, 8192, "%s/.urb", dir_c);
+  if ( c3_mkdir(ful_c, 0700) ) {
+    if ( EEXIST != errno ) {
+      fprintf(stderr, "loom: .urb create: %s\r\n", strerror(errno));
+      exit(1);
+    }
+  }
+
+  snprintf(ful_c, 8192, "%s/.urb/chk", dir_c);
+  if ( c3_mkdir(ful_c, 0700) ) {
+    if ( EEXIST != errno ) {
+      fprintf(stderr, "loom: .urb/chk create: %s\r\n", strerror(errno));
+      exit(1);
+    }
+  }
+
+  return strdup(dir_c);
+}
+
 /* u3m_boot(): start the u3 system. return next event, starting from 1.
 */
 c3_d
@@ -2083,7 +2119,7 @@ u3m_boot(c3_c* dir_c, size_t len_i)
 
   /* Activate the storage system.
   */
-  nuu_o = u3e_live(c3n, dir_c);
+  nuu_o = u3e_live(c3n, u3m_pier(dir_c));
 
   /* Activate tracing.
   */
diff --git a/pkg/noun/manage.h b/pkg/noun/manage.h
index 297585c486..56d4ca7901 100644
--- a/pkg/noun/manage.h
+++ b/pkg/noun/manage.h
@@ -14,6 +14,11 @@
         c3_d
         u3m_boot(c3_c* dir_c, size_t len_i);
 
+      /* u3m_pier(): make a pier.
+      */
+        c3_c*
+        u3m_pier(c3_c* dir_c);
+
       /* u3m_boot_lite(): start without checkpointing.
       */
         c3_d
diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 4d95a9516e..2085262bd7 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -1028,10 +1028,6 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
     strcat(dir_c, "/.urb/get");
     c3_mkdir(dir_c, 0700);
 
-    strcpy(dir_c, pax_c);
-    strcat(dir_c, "/.urb/chk");
-    c3_mkdir(dir_c, 0700);
-
     c3_free(dir_c);
   }
 

From 8668d732bbed6896d800e6bccf12687d47160d1d Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Thu, 29 Jun 2023 21:11:38 -0400
Subject: [PATCH 48/59] cli: subcommands auto-migrate disk format

---
 pkg/vere/main.c | 37 +++++++++++--------------------------
 1 file changed, 11 insertions(+), 26 deletions(-)

diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index 0cc3b4afc8..2291ed3b4f 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -1561,11 +1561,11 @@ _cw_info(c3_i argc, c3_c* argv[])
     exit(1);
   }
 
-  c3_d     eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
+  u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
   u3_disk* log_u = _cw_disk_init(u3_Host.dir_c);
 
   fprintf(stderr, "\r\nurbit: %s at event %" PRIu64 "\r\n",
-                  u3_Host.dir_c, eve_d);
+                  u3_Host.dir_c, u3_Host.eve_d);
 
   u3_disk_slog(log_u);
   printf("\n");
@@ -1720,17 +1720,17 @@ _cw_cram(c3_i argc, c3_c* argv[])
     exit(1);
   }
 
-  c3_d     eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
+  u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
   u3_disk* log_u = _cw_disk_init(u3_Host.dir_c); // XX s/b try_aquire lock
   c3_o  ret_o;
 
   fprintf(stderr, "urbit: cram: preparing\r\n");
 
-  if ( c3n == (ret_o = u3u_cram(u3_Host.dir_c, eve_d)) ) {
+  if ( c3n == (ret_o = u3u_cram(u3_Host.dir_c, u3_Host.eve_d)) ) {
     fprintf(stderr, "urbit: cram: unable to jam state\r\n");
   }
   else {
-    fprintf(stderr, "urbit: cram: rock saved at event %" PRIu64 "\r\n", eve_d);
+    fprintf(stderr, "urbit: cram: rock saved at event %" PRIu64 "\r\n", u3_Host.eve_d);
   }
 
   //  save even on failure, as we just did all the work of deduplication
@@ -1832,12 +1832,11 @@ _cw_queu(c3_i argc, c3_c* argv[])
     exit(1);
   }
   else {
+    u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
     u3_disk* log_u = _cw_disk_init(u3_Host.dir_c); // XX s/b try_aquire lock
 
     fprintf(stderr, "urbit: queu: preparing\r\n");
 
-    u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
-
     //  XX can spuriously fail do to corrupt memory-image checkpoint,
     //  need a u3m_half_boot equivalent
     //  workaround is to delete/move the checkpoint in case of corruption
@@ -1925,11 +1924,11 @@ _cw_meld(c3_i argc, c3_c* argv[])
     exit(1);
   }
 
+  u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
   u3_disk* log_u = _cw_disk_init(u3_Host.dir_c); // XX s/b try_aquire lock
   c3_w     pre_w;
 
   u3C.wag_w |= u3o_hashless;
-  u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
 
   pre_w = u3a_open(u3R);
   u3u_meld();
@@ -2090,9 +2089,9 @@ _cw_pack(c3_i argc, c3_c* argv[])
     exit(1);
   }
 
+  u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
   u3_disk* log_u = _cw_disk_init(u3_Host.dir_c); // XX s/b try_aquire lock
 
-  u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
   u3a_print_memory(stderr, "urbit: pack: gained", u3m_pack());
 
   u3m_save();
@@ -2246,6 +2245,7 @@ _cw_play(c3_i argc, c3_c* argv[])
 
   //  XX handle SIGTSTP so that the lockfile is not orphaned?
   //
+  u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
   u3_disk* log_u = _cw_disk_init(u3_Host.dir_c); // XX s/b try_aquire lock
 
   //  Handle SIGTSTP as if it was SIGINT.
@@ -2269,8 +2269,6 @@ _cw_play(c3_i argc, c3_c* argv[])
     _cw_play_snap(log_u);
   }
 
-  u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
-
   u3C.slog_f = _cw_play_slog;
 
   {
@@ -2435,19 +2433,9 @@ _cw_chop(c3_i argc, c3_c* argv[])
   }
 
   // gracefully shutdown the pier if it's running
+  u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
   u3_disk* log_u = _cw_disk_init(u3_Host.dir_c);
 
-  // note: this includes patch applications (if any)
-  u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
-
-  // check if there's a *current* snapshot
-  if ( log_u->dun_d != u3A->eve_d ) {
-    fprintf(stderr, "chop: error: snapshot is out of date, please "
-                    "start/shutdown your pier gracefully first\r\n");
-    fprintf(stderr, "chop: eve_d: %" PRIc3_d ", dun_d: %" PRIc3_d "\r\n", u3A->eve_d, log_u->dun_d);
-    exit(1);
-  }
-
   //  get latest epoch number prior to creating a new one
   c3_d pre_d;
   if ( c3n == u3_disk_epoc_last(log_u, &pre_d) ) {
@@ -2523,7 +2511,6 @@ _cw_chop(c3_i argc, c3_c* argv[])
   // cleanup
   u3_dire_free(ned_u);
   u3_disk_exit(log_u);
-  u3m_stop();
 
   // success
   fprintf(stderr, "chop: event log truncation complete\r\n");
@@ -2580,11 +2567,9 @@ _cw_roll(c3_i argc, c3_c* argv[])
   }
 
   // gracefully shutdown the pier if it's running
+  u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
   u3_disk* log_u = _cw_disk_init(u3_Host.dir_c);
 
-  // note: this includes patch applications (if any)
-  u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
-
   // check if there's a *current* snapshot
   if ( log_u->dun_d != u3A->eve_d ) {
     fprintf(stderr, "roll: error: snapshot is out of date, please "

From 5e3f25358bad01beb63f50bf5aaca5d02c3bd403 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Sun, 2 Jul 2023 07:57:06 -0400
Subject: [PATCH 49/59] cli: fix `play` boot

---
 pkg/vere/main.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index 2291ed3b4f..fd7f6703d4 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -342,7 +342,7 @@ _main_getopt(c3_i argc, c3_c** argv)
           return c3n;
         } else {
           u3_Host.ops_u.sap_w = arg_w * 60;
-          if ( 0 == u3_Host.ops_u.sap_w)
+          if ( 0 == u3_Host.ops_u.sap_w )
             return c3n;
         }
         break;
@@ -1197,7 +1197,7 @@ _cw_disk_init(c3_c* dir_c)
   u3_disk*  log_u = u3_disk_init(dir_c, cb_u);
 
   if ( !log_u ) {
-    fprintf(stderr, "unable to open event log\n");;
+    fprintf(stderr, "unable to open event log\n");
     exit(1);
   }
 
@@ -1924,12 +1924,12 @@ _cw_meld(c3_i argc, c3_c* argv[])
     exit(1);
   }
 
+  u3C.wag_w |= u3o_hashless;
+
   u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
   u3_disk* log_u = _cw_disk_init(u3_Host.dir_c); // XX s/b try_aquire lock
   c3_w     pre_w;
 
-  u3C.wag_w |= u3o_hashless;
-
   pre_w = u3a_open(u3R);
   u3u_meld();
   u3a_print_memory(stderr, "urbit: meld: gained", u3u_meld());
@@ -2245,7 +2245,6 @@ _cw_play(c3_i argc, c3_c* argv[])
 
   //  XX handle SIGTSTP so that the lockfile is not orphaned?
   //
-  u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
   u3_disk* log_u = _cw_disk_init(u3_Host.dir_c); // XX s/b try_aquire lock
 
   //  Handle SIGTSTP as if it was SIGINT.
@@ -2269,6 +2268,8 @@ _cw_play(c3_i argc, c3_c* argv[])
     _cw_play_snap(log_u);
   }
 
+  u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
+
   u3C.slog_f = _cw_play_slog;
 
   {

From 1140c4e0da4fff6c13ec3a83c1b28342c092d175 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Wed, 5 Jul 2023 14:55:42 -0400
Subject: [PATCH 50/59] play: support full replay of unmigrated piers without
 valid snapshot

---
 pkg/vere/disk.c | 20 ++++++++++++++++----
 pkg/vere/main.c | 14 +++++++++++---
 2 files changed, 27 insertions(+), 7 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 2085262bd7..56f9a81a30 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -1043,11 +1043,23 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
       return 0;
     }
 
-    //  migrate to the correct disk format if necessary
+    //  try migrating the disk format
     if ( c3n == u3_disk_migrate(log_u) ) {
-      fprintf(stderr, "disk: failed to migrate to v%d\r\n", U3D_VER1);
-      c3_free(log_u);
-      return 0;
+      fprintf(stderr, "disk: loading old format\r\n");
+
+      if ( 0 == (log_u->mdb_u = u3_lmdb_init(log_c, siz_i)) ) {
+        fprintf(stderr, "disk: failed to initialize lmdb\r\n");
+        c3_free(log_u);
+      }
+
+      c3_d fir_d;
+      if ( c3n == u3_lmdb_gulf(log_u->mdb_u, &fir_d, &log_u->dun_d) ) {
+        fprintf(stderr, "disk: failed to load latest event from lmdb\r\n");
+        c3_free(log_u);
+        return 0;
+      }
+
+      return log_u;
     }
 
     //  get latest epoch number
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index fd7f6703d4..36cdd6c456 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -1924,12 +1924,12 @@ _cw_meld(c3_i argc, c3_c* argv[])
     exit(1);
   }
 
-  u3C.wag_w |= u3o_hashless;
-
   u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
   u3_disk* log_u = _cw_disk_init(u3_Host.dir_c); // XX s/b try_aquire lock
   c3_w     pre_w;
 
+  u3C.wag_w |= u3o_hashless;
+
   pre_w = u3a_open(u3R);
   u3u_meld();
   u3a_print_memory(stderr, "urbit: meld: gained", u3u_meld());
@@ -2245,7 +2245,11 @@ _cw_play(c3_i argc, c3_c* argv[])
 
   //  XX handle SIGTSTP so that the lockfile is not orphaned?
   //
-  u3_disk* log_u = _cw_disk_init(u3_Host.dir_c); // XX s/b try_aquire lock
+  u3_disk* log_u;
+  if ( 0 == (log_u = _cw_disk_init(u3_Host.dir_c)) ) {
+    fprintf(stderr, "mars: failed to load event log\r\n");
+    exit(1);
+  }
 
   //  Handle SIGTSTP as if it was SIGINT.
   //
@@ -2281,6 +2285,10 @@ _cw_play(c3_i argc, c3_c* argv[])
     };
 
     u3_mars_play(&mar_u, eve_d, sap_d);
+
+    //  migrate after replay, if necessary
+    u3_Host.eve_d = mar_u.dun_d;
+    u3_disk_migrate(log_u);
   }
 
   u3_disk_exit(log_u);

From fcd9bcc74c59a8d2f0c6762ed4be24dab12e9757 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Thu, 6 Jul 2023 17:09:42 -0400
Subject: [PATCH 51/59] always replay synchronously on restart

---
 pkg/vere/disk.c |  71 +++++++++++++++++++----------
 pkg/vere/main.c | 119 +++++++++++++++++++++++++++---------------------
 pkg/vere/pier.c |   2 +-
 pkg/vere/vere.h |   8 +++-
 4 files changed, 120 insertions(+), 80 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index 56f9a81a30..cc07da2827 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -975,7 +975,7 @@ u3_disk_slog(u3_disk* log_u)
 /* u3_disk_init(): load or create pier directories and event log.
 */
 u3_disk*
-u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
+u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u, c3_o mig_o)
 {
   u3_disk* log_u = c3_calloc(sizeof(*log_u));
   log_u->liv_o = c3n;
@@ -1043,23 +1043,33 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u)
       return 0;
     }
 
-    //  try migrating the disk format
-    if ( c3n == u3_disk_migrate(log_u) ) {
-      fprintf(stderr, "disk: loading old format\r\n");
-
-      if ( 0 == (log_u->mdb_u = u3_lmdb_init(log_c, siz_i)) ) {
-        fprintf(stderr, "disk: failed to initialize lmdb\r\n");
-        c3_free(log_u);
+    if ( c3y == u3_disk_need_migrate(log_u) ) {
+      if ( c3y == mig_o ) {
+        if ( c3n == u3_disk_migrate(log_u) ) {
+          fprintf(stderr, "disk: failed to migrate log\r\n");
+          c3_free(log_u);
+          return 0;
+        }
       }
+      else {
+        fprintf(stderr, "disk: loading old format\r\n");
 
-      c3_d fir_d;
-      if ( c3n == u3_lmdb_gulf(log_u->mdb_u, &fir_d, &log_u->dun_d) ) {
-        fprintf(stderr, "disk: failed to load latest event from lmdb\r\n");
-        c3_free(log_u);
-        return 0;
-      }
+        if ( 0 == (log_u->mdb_u = u3_lmdb_init(log_c, siz_i)) ) {
+          fprintf(stderr, "disk: failed to initialize lmdb\r\n");
+          c3_free(log_u);
+        }
 
-      return log_u;
+        c3_d fir_d;
+        if ( c3n == u3_lmdb_gulf(log_u->mdb_u, &fir_d, &log_u->dun_d) ) {
+          fprintf(stderr, "disk: failed to load latest event from lmdb\r\n");
+          c3_free(log_u);
+          return 0;
+        }
+
+        log_u->sen_d = log_u->dun_d;
+
+        return log_u;
+      }
     }
 
     //  get latest epoch number
@@ -1291,9 +1301,27 @@ u3_disk_epoc_vere(u3_disk* log_u, c3_d epo_d, c3_c* ver_w)
   return c3y;
 }
 
+/* u3_disk_need_migrate: does the desk need to be migrated?
+*/
+c3_o
+u3_disk_need_migrate(u3_disk* log_u)
+{
+  //  check if data.mdb is readable in log directory
+  c3_c dut_c[8193];
+  snprintf(dut_c, sizeof(dut_c), "%s/data.mdb", log_u->com_u->pax_c);
+  if ( !_(u3_Host.ops_u.nuu)
+       && 0 != access(dut_c, R_OK) ) {
+    // if .urb/log/data.mdb is not readable, skip migration
+    return c3n;
+  }
+
+  return c3y;
+}
+
 /* u3_disk_migrate: migrates disk format.
  */
-c3_o u3_disk_migrate(u3_disk* log_u)
+c3_o
+u3_disk_migrate(u3_disk* log_u)
 {
   /*  migration steps:
    *  0. detect whether we need to migrate or not
@@ -1315,14 +1343,6 @@ c3_o u3_disk_migrate(u3_disk* log_u)
    *  5. delete old data.mdb and lock.mdb files (c3_unlink() calls)
    */
 
-  //  check if data.mdb is readable in log directory
-  c3_c dut_c[8193];
-  snprintf(dut_c, sizeof(dut_c), "%s/data.mdb", log_u->com_u->pax_c);
-  if ( !_(u3_Host.ops_u.nuu)
-       && 0 != access(dut_c, R_OK) ) {
-    // if .urb/log/data.mdb is not readable, skip migration
-    return c3y;
-  }
 
   //  check if lock.mdb is readable in log directory
   c3_o luk_o = c3n;
@@ -1378,8 +1398,9 @@ c3_o u3_disk_migrate(u3_disk* log_u)
   }
 
   //  create hard links to data.mdb and lock.mdb in 0i0/
-  c3_c epo_c[8193], dat_c[8193], lok_c[8193];
+  c3_c epo_c[8193], dut_c[8193], dat_c[8193], lok_c[8193];
   snprintf(epo_c, sizeof(epo_c), "%s/0i0", log_u->com_u->pax_c);
+  snprintf(dut_c, sizeof(dut_c), "%s/data.mdb", log_u->com_u->pax_c);
   snprintf(dat_c, sizeof(dat_c), "%s/data.mdb", epo_c);
   snprintf(lok_c, sizeof(lok_c), "%s/lock.mdb", epo_c);
 
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index 36cdd6c456..abda7a5f14 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -1194,7 +1194,7 @@ static u3_disk*
 _cw_disk_init(c3_c* dir_c)
 {
   u3_disk_cb cb_u = {0};
-  u3_disk*  log_u = u3_disk_init(dir_c, cb_u);
+  u3_disk*  log_u = u3_disk_init(dir_c, cb_u, c3y);
 
   if ( !log_u ) {
     fprintf(stderr, "unable to open event log\n");
@@ -2151,6 +2151,69 @@ _cw_play_exit(c3_i int_i)
   raise(SIGINT);
 }
 
+/* _cw_play_impl(): replay events, but better.
+*/
+static void
+_cw_play_impl(c3_d eve_d, c3_d sap_d, c3_o mel_o, c3_o sof_o, c3_o ful_o)
+{
+  //  XX factor this into its own function for calling here and
+  //     from main() (urbit play and boot replay)
+
+  //  XX handle SIGTSTP so that the lockfile is not orphaned?
+  //
+  u3_disk* log_u;
+  if ( 0 == (log_u = u3_disk_init(u3_Host.dir_c, (u3_disk_cb){0}, c3n)) ) {
+    fprintf(stderr, "mars: failed to load event log\r\n");
+    exit(1);
+  }
+
+  //  Handle SIGTSTP as if it was SIGINT.
+  //
+  //    Configured here using signal() so as to be immediately available.
+  //
+  signal(SIGTSTP, _cw_play_exit);
+
+  //  XX source these from a shared struct ops_u
+  if ( c3y == mel_o ) {
+    u3C.wag_w |= u3o_auto_meld;
+  }
+
+  if ( c3y == sof_o ) {
+    u3C.wag_w |= u3o_soft_mugs;
+  }
+
+  u3C.wag_w |= u3o_hashless;
+
+  if ( c3y == ful_o ) {
+    u3l_log("mars: preparing for full replay");
+    _cw_play_snap(log_u);
+  }
+
+  u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
+
+  u3C.slog_f = _cw_play_slog;
+
+  {
+    u3_mars mar_u = {
+      .log_u = log_u,
+      .dir_c = u3_Host.dir_c,
+      .sen_d = u3A->eve_d,
+      .dun_d = u3A->eve_d,
+    };
+
+    u3_mars_play(&mar_u, eve_d, sap_d);
+
+    //  migrate after replay, if necessary
+    u3_Host.eve_d = mar_u.dun_d;
+    if ( c3y == u3_disk_need_migrate(log_u) ) {
+      u3_disk_migrate(log_u);
+    }
+  }
+
+  u3_disk_exit(log_u);
+  u3m_stop();
+}
+
 /* _cw_play(): replay events, but better.
 */
 static void
@@ -2243,56 +2306,7 @@ _cw_play(c3_i argc, c3_c* argv[])
     exit(1);
   }
 
-  //  XX handle SIGTSTP so that the lockfile is not orphaned?
-  //
-  u3_disk* log_u;
-  if ( 0 == (log_u = _cw_disk_init(u3_Host.dir_c)) ) {
-    fprintf(stderr, "mars: failed to load event log\r\n");
-    exit(1);
-  }
-
-  //  Handle SIGTSTP as if it was SIGINT.
-  //
-  //    Configured here using signal() so as to be immediately available.
-  //
-  signal(SIGTSTP, _cw_play_exit);
-
-  if ( c3y == mel_o ) {
-    u3C.wag_w |= u3o_auto_meld;
-  }
-
-  if ( c3y == sof_o ) {
-    u3C.wag_w |= u3o_soft_mugs;
-  }
-
-  u3C.wag_w |= u3o_hashless;
-
-  if ( c3y == ful_o ) {
-    u3l_log("mars: preparing for full replay");
-    _cw_play_snap(log_u);
-  }
-
-  u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
-
-  u3C.slog_f = _cw_play_slog;
-
-  {
-    u3_mars mar_u = {
-      .log_u = log_u,
-      .dir_c = u3_Host.dir_c,
-      .sen_d = u3A->eve_d,
-      .dun_d = u3A->eve_d,
-    };
-
-    u3_mars_play(&mar_u, eve_d, sap_d);
-
-    //  migrate after replay, if necessary
-    u3_Host.eve_d = mar_u.dun_d;
-    u3_disk_migrate(log_u);
-  }
-
-  u3_disk_exit(log_u);
-  u3m_stop();
+  _cw_play_impl(eve_d, sap_d, mel_o, sof_o, ful_o);
 }
 
 /* _cw_prep(): prepare for upgrade
@@ -3085,8 +3099,7 @@ main(c3_i   argc,
     //  we need the current snapshot's latest event number to
     //  validate whether we can execute disk migration
     if ( u3_Host.ops_u.nuu == c3n ) {
-      u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
-      u3m_stop();
+      _cw_play_impl(0, 0, c3n, c3n, c3n);
       //  XX  unmap loom, else parts of the snapshot could be left in memory
     }
 
diff --git a/pkg/vere/pier.c b/pkg/vere/pier.c
index c2e7c47026..81e744e4f7 100644
--- a/pkg/vere/pier.c
+++ b/pkg/vere/pier.c
@@ -1643,7 +1643,7 @@ _pier_init(c3_w wag_w, c3_c* pax_c)
       .write_bail_f = _pier_on_disk_write_bail
     };
 
-    if ( !(pir_u->log_u = u3_disk_init(pax_c, cb_u)) ) {
+    if ( !(pir_u->log_u = u3_disk_init(pax_c, cb_u, c3n)) ) {
       c3_free(pir_u);
       return 0;
     }
diff --git a/pkg/vere/vere.h b/pkg/vere/vere.h
index c72417f86d..9cb9e88d8b 100644
--- a/pkg/vere/vere.h
+++ b/pkg/vere/vere.h
@@ -8,6 +8,7 @@
 #include "noun.h"
 #include "serf.h"
 #include "uv.h"
+#include <types.h>
 
   /** Quasi-tunable parameters.
   **/
@@ -935,7 +936,7 @@
       /* u3_disk_init(): load or create pier directories and event log.
       */
         u3_disk*
-        u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u);
+        u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u, c3_o mig_o);
 
       /* u3_disk_etch(): serialize an event for persistence. RETAIN [eve]
       */
@@ -1030,6 +1031,11 @@
        c3_o
        u3_disk_vere_diff(u3_disk* log_u);
 
+      /* u3_disk_need_migrate(): does the disk need migration?
+      */
+        c3_o
+        u3_disk_need_migrate(u3_disk* log_u);
+
       /* u3_disk_migrate(): migrates disk format.
        */
         c3_o

From f229c4db9542fc104c3ff527add24a8ac0031fe0 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Thu, 6 Jul 2023 17:13:16 -0400
Subject: [PATCH 52/59] deprecated async replay codepath

---
 pkg/vere/pier.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/pkg/vere/pier.c b/pkg/vere/pier.c
index 81e744e4f7..44e9ee4d35 100644
--- a/pkg/vere/pier.c
+++ b/pkg/vere/pier.c
@@ -1408,6 +1408,8 @@ _pier_on_lord_live(void* ptr_v)
       //  XX print bootstrap commit complete
       //  XX s/b boot_complete_cb
       //
+      //  XX this codepath should never be hit due to sync replay
+      u3l_log("pier: warning: async replay");
       _pier_play_init(pir_u, log_u->dun_d);
     }
   }

From 3c8c0b219d3d926a32480149acdf37c8950e9406 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Wed, 12 Jul 2023 12:05:50 -0400
Subject: [PATCH 53/59] disk: fix migrate scenarios

---
 pkg/vere/disk.c | 4 ++--
 pkg/vere/main.c | 3 ---
 pkg/vere/pier.c | 2 +-
 3 files changed, 3 insertions(+), 6 deletions(-)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index cc07da2827..f87ce92154 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -1310,8 +1310,8 @@ u3_disk_need_migrate(u3_disk* log_u)
   c3_c dut_c[8193];
   snprintf(dut_c, sizeof(dut_c), "%s/data.mdb", log_u->com_u->pax_c);
   if ( !_(u3_Host.ops_u.nuu)
-       && 0 != access(dut_c, R_OK) ) {
-    // if .urb/log/data.mdb is not readable, skip migration
+       && 0 != access(dut_c, F_OK) ) {
+    // if .urb/log/data.mdb does not exist, skip migration
     return c3n;
   }
 
diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index 5cc44e0cfa..b09ccdd346 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -2153,9 +2153,6 @@ _cw_play_exit(c3_i int_i)
 static void
 _cw_play_impl(c3_d eve_d, c3_d sap_d, c3_o mel_o, c3_o sof_o, c3_o ful_o)
 {
-  //  XX factor this into its own function for calling here and
-  //     from main() (urbit play and boot replay)
-
   //  XX handle SIGTSTP so that the lockfile is not orphaned?
   //
   u3_disk* log_u;
diff --git a/pkg/vere/pier.c b/pkg/vere/pier.c
index 44e9ee4d35..830a2c1e49 100644
--- a/pkg/vere/pier.c
+++ b/pkg/vere/pier.c
@@ -1645,7 +1645,7 @@ _pier_init(c3_w wag_w, c3_c* pax_c)
       .write_bail_f = _pier_on_disk_write_bail
     };
 
-    if ( !(pir_u->log_u = u3_disk_init(pax_c, cb_u, c3n)) ) {
+    if ( !(pir_u->log_u = u3_disk_init(pax_c, cb_u, c3y)) ) {
       c3_free(pir_u);
       return 0;
     }

From 1556567b3333b3af4847753d239e0662e06f5c39 Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Tue, 18 Jul 2023 12:03:44 -0400
Subject: [PATCH 54/59] disk/events: simplify code

---
 pkg/noun/events.c |  8 ++++----
 pkg/vere/disk.c   | 10 ++++------
 2 files changed, 8 insertions(+), 10 deletions(-)

diff --git a/pkg/noun/events.c b/pkg/noun/events.c
index 8b8d98993e..0294f522ab 100644
--- a/pkg/noun/events.c
+++ b/pkg/noun/events.c
@@ -1344,12 +1344,12 @@ c3_o
 u3e_backup(c3_c* pux_c, c3_c* pax_c, c3_o ovw_o)
 {
   //  source image files from [pux_c]
-  u3e_image nux_u = { .nam_c = "north" };
-  u3e_image sux_u = { .nam_c = "south" };
+  u3e_image nux_u = { .nam_c = "north", .pgs_w = 0 };
+  u3e_image sux_u = { .nam_c = "south", .pgs_w = 0 };
 
   //  destination image files to [pax_c]
-  u3e_image nax_u = { .nam_c = "north" };
-  u3e_image sax_u = { .nam_c = "south" };
+  u3e_image nax_u = { .nam_c = "north", .pgs_w = 0 };
+  u3e_image sax_u = { .nam_c = "south", .pgs_w = 0 };
 
   c3_i mod_i = O_RDWR | O_CREAT;
 
diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index f87ce92154..e506b71153 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -1044,12 +1044,10 @@ u3_disk_init(c3_c* pax_c, u3_disk_cb cb_u, c3_o mig_o)
     }
 
     if ( c3y == u3_disk_need_migrate(log_u) ) {
-      if ( c3y == mig_o ) {
-        if ( c3n == u3_disk_migrate(log_u) ) {
-          fprintf(stderr, "disk: failed to migrate log\r\n");
-          c3_free(log_u);
-          return 0;
-        }
+      if ( (c3y == mig_o) && (c3n == u3_disk_migrate(log_u)) ) {
+        fprintf(stderr, "disk: failed to migrate log\r\n");
+        c3_free(log_u);
+        return 0;
       }
       else {
         fprintf(stderr, "disk: loading old format\r\n");

From f9a2d315c3354ca25d02493f48f5d67feed5d1ad Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Wed, 26 Jul 2023 10:20:09 -0400
Subject: [PATCH 55/59] disk: add "migration starting" printf

---
 pkg/vere/disk.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/pkg/vere/disk.c b/pkg/vere/disk.c
index e506b71153..f24c11e11b 100644
--- a/pkg/vere/disk.c
+++ b/pkg/vere/disk.c
@@ -1341,6 +1341,7 @@ u3_disk_migrate(u3_disk* log_u)
    *  5. delete old data.mdb and lock.mdb files (c3_unlink() calls)
    */
 
+  fprintf(stderr, "disk: migrating disk to v%d format\r\n", U3D_VER1);
 
   //  check if lock.mdb is readable in log directory
   c3_o luk_o = c3n;

From ea6f278bdf02b60d795489be04f35d471181769d Mon Sep 17 00:00:00 2001
From: Matthew LeVan <matthew@coeli.network>
Date: Wed, 16 Aug 2023 08:40:56 -0600
Subject: [PATCH 56/59] disk: move `u3C.wag_w` assignment before boot call

---
 pkg/vere/main.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/pkg/vere/main.c b/pkg/vere/main.c
index b09ccdd346..cca3081a9b 100644
--- a/pkg/vere/main.c
+++ b/pkg/vere/main.c
@@ -1924,11 +1924,11 @@ _cw_meld(c3_i argc, c3_c* argv[])
     exit(1);
   }
 
+  u3C.wag_w |= u3o_hashless;
+
   u3_Host.eve_d = u3m_boot(u3_Host.dir_c, (size_t)1 << u3_Host.ops_u.lom_y);
   u3_disk* log_u = _cw_disk_init(u3_Host.dir_c); // XX s/b try_aquire lock
 
-  u3C.wag_w |= u3o_hashless;
-
   u3a_print_memory(stderr, "urbit: meld: gained", u3u_meld());
 
   u3m_save();

From 53a4664b75e7eb8390e0f2287e641ef3795f4050 Mon Sep 17 00:00:00 2001
From: Midlyx Hatrys <galactic.koreiko@gmail.com>
Date: Fri, 5 May 2023 13:08:49 +0300
Subject: [PATCH 57/59] flake: update

Otherwise devenv fails to work on newer NixOS'es, looks like because
th glibc is too old.
---
 flake.lock | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/flake.lock b/flake.lock
index 12ffaf6f77..45c4d968fb 100644
--- a/flake.lock
+++ b/flake.lock
@@ -2,11 +2,11 @@
   "nodes": {
     "nixpkgs": {
       "locked": {
-        "lastModified": 1677383253,
-        "narHash": "sha256-UfpzWfSxkfXHnb4boXZNaKsAcUrZT9Hw+tao1oZxd08=",
+        "lastModified": 1683236849,
+        "narHash": "sha256-Y7PNBVLOBvZrmrFmHgXUBUA1lM72tl6JGIn1trOeuyE=",
         "owner": "nixos",
         "repo": "nixpkgs",
-        "rev": "9952d6bc395f5841262b006fbace8dd7e143b634",
+        "rev": "374ffe54403c3c42d97a513ac7a14ce1b5b86e30",
         "type": "github"
       },
       "original": {
@@ -23,11 +23,11 @@
         ]
       },
       "locked": {
-        "lastModified": 1675933616,
-        "narHash": "sha256-/rczJkJHtx16IFxMmAWu5nNYcSXNg1YYXTHoGjLrLUA=",
+        "lastModified": 1682984683,
+        "narHash": "sha256-fSMthG+tp60AHhNmaHc4StT3ltfHkQsJtN8GhfLWmtI=",
         "owner": "hercules-ci",
         "repo": "flake-parts",
-        "rev": "47478a4a003e745402acf63be7f9a092d51b83d7",
+        "rev": "86684881e184f41aa322e653880e497b66429f3e",
         "type": "github"
       },
       "original": {

From 28f18498129173196158c18463e9320b71f2d36c Mon Sep 17 00:00:00 2001
From: Pyry Kovanen <pkova@users.noreply.github.com>
Date: Mon, 18 Sep 2023 17:00:34 +0300
Subject: [PATCH 58/59] Update VERSION to 3.0

---
 VERSION | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/VERSION b/VERSION
index 3e162f02ea..9f55b2ccb5 100644
--- a/VERSION
+++ b/VERSION
@@ -1 +1 @@
-2.12
+3.0

From 56fce75730a71a98f24bcc039934c9d32185e76b Mon Sep 17 00:00:00 2001
From: Amadeo Bellotti <amadeo.bellotti@gmail.com>
Date: Fri, 29 Sep 2023 10:52:36 -0400
Subject: [PATCH 59/59] fixed next/kelvin workflow to not upload to docker

---
 .github/workflows/next.yml | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/.github/workflows/next.yml b/.github/workflows/next.yml
index e0dd8feeec..07250449e3 100644
--- a/.github/workflows/next.yml
+++ b/.github/workflows/next.yml
@@ -20,7 +20,3 @@ jobs:
       upload: true
       next: ${{ github.ref_name }}
     secrets: inherit
-
-  docker:
-    uses: ./.github/workflows/docker-shared.yml
-    secrets: inherit