Keeping master up to date

.gitignore

@@ -11,6 +11,7 @@ resources/playbook/group_vars/
 
 # any log files
 *.log
+log/
 
 # Byte-compiled / optimized / DLL files
 __pycache__/

bibigrid.sh

@@ -1 +1,2 @@
+#!/bin/bash
 python3 -m bibigrid.core.startup "$@"

bibigrid.yml

@@ -1,4 +1,5 @@
   # See https://cloud.denbi.de/wiki/Tutorials/BiBiGrid/ (after update)
+  # See https://github.com/BiBiServ/bibigrid/blob/master/documentation/markdown/features/configuration.md
   # First configuration also holds general cluster information and must include the master.
   # All other configurations mustn't include another master, but exactly one vpngtw instead (keys like master).
 
@@ -8,9 +9,10 @@
   # -- BEGIN: GENERAL CLUSTER INFORMATION --
   ## sshPublicKeyFiles listed here will be added to access the cluster. A temporary key is created by bibigrid itself.
   #sshPublicKeyFiles:
-  #  - [key one]
+  #  - [public key one]
 
   ## Volumes and snapshots that will be mounted to master
+  # autoMount: False # WARNING: will overwrite unidentified filesystems
   #masterMounts: # KEY NOT FULLY IMPLEMENTED YET
   #  - [mount one]
 
@@ -36,13 +38,15 @@
 
   ## Uncomment if you don't want assign a public ip to the master; for internal cluster (Tuebingen).
   #useMasterWithPublicIp: False # defaults True if False no public-ip (floating-ip) will be allocated
+  # deleteTmpKeypairAfter: False
+  # dontUploadCredentials: False
 
   # Other keys - default False
   #localFS: True
   #localDNSlookup: True
   #zabbix: True
   #nfs: True
-  #ide: True # Very useful to set on True. Use `./bibigrid.sh -i [path-to-bibigrid.yml] -ide -cid [cluster-id]` to start port forwarding to access the ide.
+  #ide: True # A nice way to view your cluster as if you were using Visual Studio Code
 
   useMasterAsCompute: True # Currently ignored by slurm
 
@@ -52,15 +56,17 @@
   # master configuration
   masterInstance:
     type: # existing type/flavor on your cloud. See launch instance>flavor for options
-    image: # existing image on your cloud. See https://openstack.cebitec.uni-bielefeld.de/project/images pick an active one. Currently only ubuntu22.04 is supported
+    image: # existing active image on your cloud. Consider using regex to prevent image updates from breaking your running cluster
     # features: # list
 
   # -- END: GENERAL CLUSTER INFORMATION --
 
+  # fallbackOnOtherImage: False # if True, most similar image by name will be picked. A regex can also be given instead.
+
   # worker configuration
   #workerInstances:
   #  - type: # existing type/flavor on your cloud. See launch instance>flavor for options
-  #    image: # same as master
+  #    image: # same as master. Consider using regex to prevent image updates from breaking your running cluster
   #    count: # any number of workers you would like to create with set type, image combination
   #    # features: # list
 
@@ -87,6 +93,10 @@
 
   # Depends on cloud site and project
   subnet: # existing subnet on your cloud. See https://openstack.cebitec.uni-bielefeld.de/project/networks/
+  # or network:
+  # gateway: # if you want to use a gateway for create.
+    # ip: # IP of gateway to use
+    # portFunction: 30000 + oct4 # variables are called: oct1.oct2.oct3.oct4
 
   # Uncomment if no full DNS service for started instances is available.
   # Currently, the case in Berlin, DKFZ, Heidelberg and Tuebingen.

bibigrid/core/actions/check.py

@@ -1,21 +1,19 @@
 """
 Module that acts as a wrapper and uses validate_configuration to validate given configuration
 """
-import logging
 from bibigrid.core.utility import validate_configuration
 
-LOG = logging.getLogger("bibigrid")
 
-
-def check(configurations, providers):
+def check(configurations, providers, log):
     """
     Uses validate_configuration to validate given configuration.
     :param configurations: list of configurations (dicts)
     :param providers: list of providers
+    :param log:
     :return:
     """
-    success = validate_configuration.ValidateConfiguration(configurations, providers).validate()
+    success = validate_configuration.ValidateConfiguration(configurations, providers, log).validate()
     check_result = "succeeded! Cluster is ready to start." if success else "failed!"
-    print(f"Total check {check_result}")
-    LOG.info("Total check returned %s.", success)
+    log.log(42, f"Total check {check_result}")
+    log.info("Total check returned %s.", success)
     return 0

bibigrid/core/actions/ide.py

@@ -2,15 +2,16 @@
 This module contains methods to establish port forwarding in order to access an ide (theia).
 """
 
-import logging
 import random
 import re
 import signal
 import subprocess
 import sys
 import time
 import webbrowser
+
 import sshtunnel
+import sympy
 
 from bibigrid.core.utility.handler import cluster_ssh_handler
 
@@ -20,7 +21,7 @@
 LOCAL_BIND_ADDRESS = 9191
 MAX_JUMP = 100
 LOCALHOST = "127.0.0.1"
-LOG = logging.getLogger("bibigrid")
+
 
 
 def sigint_handler(caught_signal, frame):  # pylint: disable=unused-argument
@@ -49,37 +50,38 @@ def is_used(ip_address):
         for line in lines:
             is_open = re.match(rf'tcp.*{ip_address}:([0-9][0-9]*).*ESTABLISHED\s*$', line)
             if is_open is not None:
-                print(line)
                 ports_used.append(is_open[1])
 
 
-def ide(cluster_id, master_provider, master_configuration):
+def ide(cluster_id, master_provider, master_configuration, log):
     """
     Creates a port forwarding from LOCAL_BIND_ADDRESS to REMOTE_BIND_ADDRESS from localhost to master of specified
     cluster
     @param cluster_id: cluster_id or ip
     @param master_provider: master's provider
     @param master_configuration: master's configuration
+    @param log:
     @return:
     """
-    LOG.info("Starting port forwarding for ide")
+    log.info("Starting port forwarding for ide")
     master_ip, ssh_user, used_private_key = cluster_ssh_handler.get_ssh_connection_info(cluster_id, master_provider,
-                                                                                        master_configuration)
+                                                                                        master_configuration, log)
     used_local_bind_address = LOCAL_BIND_ADDRESS
     if master_ip and ssh_user and used_private_key:
         attempts = 0
+        if master_configuration.get("gateway"):
+            octets = {f'oct{enum + 1}': int(elem) for enum, elem in enumerate(master_ip.split("."))}
+            port = sympy.sympify(master_configuration["gateway"]["portFunction"]).subs(dict(octets))
+            gateway = (master_configuration["gateway"]["ip"], int(port))
+        else:
+            gateway = None
         while attempts < 16:
             attempts += 1
             try:
-                with sshtunnel.SSHTunnelForwarder(
-                        ssh_address_or_host=master_ip,  # Raspberry Pi in my network
-
-                        ssh_username=ssh_user,
-                        ssh_pkey=used_private_key,
-
-                        local_bind_address=(LOCALHOST, used_local_bind_address),
-                        remote_bind_address=(LOCALHOST, REMOTE_BIND_ADDRESS)
-                ) as server:
+                with sshtunnel.SSHTunnelForwarder(ssh_address_or_host=gateway or master_ip, ssh_username=ssh_user,
+                                                  ssh_pkey=used_private_key,
+                                                  local_bind_address=(LOCALHOST, used_local_bind_address),
+                                                  remote_bind_address=(LOCALHOST, REMOTE_BIND_ADDRESS)) as server:
                     print("CTRL+C to close port forwarding when you are done.")
                     with server:
                         # opens in existing window if any default program exists
@@ -88,11 +90,11 @@ def ide(cluster_id, master_provider, master_configuration):
                             time.sleep(5)
             except sshtunnel.HandlerSSHTunnelForwarderError:
                 used_local_bind_address += random.randint(1, MAX_JUMP)
-                LOG.info("Attempt: %s. Port in use... Trying new port %s", attempts, used_local_bind_address)
+                log.info("Attempt: %s. Port in use... Trying new port %s", attempts, used_local_bind_address)
     if not master_ip:
-        LOG.warning("Cluster id %s doesn't match an existing cluster with a master.", cluster_id)
+        log.warning("Cluster id %s doesn't match an existing cluster with a master.", cluster_id)
     if not ssh_user:
-        LOG.warning("No ssh user has been specified in the first configuration.")
+        log.warning("No ssh user has been specified in the first configuration.")
     if not used_private_key:
-        LOG.warning("No matching sshPublicKeyFiles can be found in the first configuration or in .bibigrid")
+        log.warning("No matching sshPublicKeyFiles can be found in the first configuration or in .bibigrid")
     return 1

bibigrid/core/actions/list_clusters.py

@@ -3,22 +3,22 @@
 This includes a method to create a dictionary containing all running clusters and their servers.
 """
 
-import logging
 import pprint
 import re
 
 from bibigrid.core.actions import create
 
 SERVER_REGEX = re.compile(r"^bibigrid-((master)-([a-zA-Z0-9]+)|(worker|vpngtw)\d+-([a-zA-Z0-9]+)-\d+)$")
-LOG = logging.getLogger("bibigrid")
 
-def dict_clusters(providers):
+
+def dict_clusters(providers, log):
     """
     Creates a dictionary containing all servers by type and provider information
     :param providers: list of all providers
+    :param log:
     :return: list of all clusters in yaml format
     """
-    LOG.info("Creating cluster dictionary...")
+    log.info("Creating cluster dictionary...")
     cluster_dict = {}
     for provider in providers:
         servers = provider.list_servers()
@@ -53,56 +53,59 @@ def setup(cluster_dict, cluster_id, server, provider):
     server["cloud_specification"] = provider.cloud_specification["identifier"]
 
 
-def print_list_clusters(cluster_id, providers):
+def log_list(cluster_id, providers, log):
     """
     Calls dict_clusters and gives a visual representation of the found cluster.
     Detail depends on whether a cluster_id is given or not.
     :param cluster_id:
     :param providers:
+    :param log:
     :return:
     """
-    cluster_dict = dict_clusters(providers=providers)
-    if cluster_id: # pylint: disable=too-many-nested-blocks
+    cluster_dict = dict_clusters(providers=providers, log=log)
+    if cluster_id:  # pylint: disable=too-many-nested-blocks
         if cluster_dict.get(cluster_id):
-            LOG.info("Printing specific cluster_dictionary")
-            master_count, worker_count, vpn_count = get_size_overview(cluster_dict[cluster_id])
-            print(f"\tCluster has {master_count} master, {vpn_count} vpngtw and {worker_count} regular workers. "
-                  f"The cluster is spread over {vpn_count + master_count} reachable provider(s).")
+            log.info("Printing specific cluster_dictionary")
+            master_count, worker_count, vpn_count = get_size_overview(cluster_dict[cluster_id], log)
+            log.log(42, f"\tCluster has {master_count} master, {vpn_count} vpngtw and {worker_count} regular workers. "
+                       f"The cluster is spread over {vpn_count + master_count} reachable provider(s).")
             pprint.pprint(cluster_dict[cluster_id])
         else:
-            LOG.info("Cluster with cluster-id {cluster_id} not found.")
-            print(f"Cluster with cluster-id {cluster_id} not found.")
+            log.info("Cluster with cluster-id {cluster_id} not found.")
+            log.log(42, f"Cluster with cluster-id {cluster_id} not found.")
     else:
-        LOG.info("Printing overview of cluster all clusters")
+        log.info("Printing overview of cluster all clusters")
         if cluster_dict:
             for cluster_key_id, cluster_node_dict in cluster_dict.items():
-                print(f"Cluster-ID: {cluster_key_id}")
+                log.log(42, f"Cluster-ID: {cluster_key_id}")
                 master = cluster_node_dict.get('master')
                 if master:
                     for key in ["name", "user_id", "launched_at", "key_name", "public_v4", "public_v6", "provider"]:
                         value = cluster_node_dict['master'].get(key)
                         if value:
-                            print(f"\t{key}: {value}")
+                            log.log(42, f"\t{key}: {value}")
                     security_groups = get_security_groups(cluster_node_dict)
-                    print(f"\tsecurity_groups: {security_groups}")
+                    log.log(42, f"\tsecurity_groups: {security_groups}")
                     networks = get_networks(cluster_node_dict)
-                    print(f"\tnetwork: {pprint.pformat(networks)}")
+                    log.log(42, f"\tnetwork: {pprint.pformat(networks)}")
                 else:
-                    LOG.warning("No master for cluster: %s.", cluster_key_id)
-                master_count, worker_count, vpn_count = get_size_overview(cluster_node_dict)
-                print(f"\tCluster has {master_count} master, {vpn_count} vpngtw and {worker_count} regular workers. "
-                      f"The cluster is spread over {vpn_count + master_count} reachable provider(s).")
+                    log.warning("No master for cluster: %s.", cluster_key_id)
+                master_count, worker_count, vpn_count = get_size_overview(cluster_node_dict, log)
+                log.log(42,
+                        f"\tCluster has {master_count} master, {vpn_count} vpngtw and {worker_count} regular workers. "
+                        f"The cluster is spread over {vpn_count + master_count} reachable provider(s).")
         else:
-            print("No cluster found.")
+            log.log(42, "No cluster found.")
     return 0
 
 
-def get_size_overview(cluster_dict):
+def get_size_overview(cluster_dict, log):
     """
     :param cluster_dict: dictionary of cluster to size_overview
+    :param log:
     :return: number of masters, number of workers, number of vpns
     """
-    LOG.info("Printing size overview")
+    log.info("Printing size overview")
     master_count = int(bool(cluster_dict.get("master")))
     worker_count = len(cluster_dict.get("workers") or "")
     vpn_count = len(cluster_dict.get("vpngtws") or "")
@@ -135,19 +138,20 @@ def get_security_groups(cluster_dict):
     return security_groups
 
 
-def get_master_access_ip(cluster_id, master_provider):
+def get_master_access_ip(cluster_id, master_provider, log):
     """
     Returns master's ip of cluster cluster_id
     :param master_provider: master's provider
     :param cluster_id: Id of cluster
+    :param log:
     :return: public ip of master
     """
-    LOG.info("Finding master ip for cluster %s...", cluster_id)
+    log.info("Finding master ip for cluster %s...", cluster_id)
     servers = master_provider.list_servers()
     for server in servers:
         master = create.MASTER_IDENTIFIER(cluster_id=cluster_id)
         if server["name"].startswith(master):
             return server.get("public_v4") or server.get("public_v6") or server.get("private_v4")
-    LOG.warning("Cluster %s not found on master_provider %s.", cluster_id,
+    log.warning("Cluster %s not found on master_provider %s.", cluster_id,
                 master_provider.cloud_specification["identifier"])
     return None