Merge branch 'distribworks:master' into master

.github/workflows/main.yml

@@ -15,7 +15,7 @@ jobs:
     - name: Set up Go
       uses: actions/setup-go@v3
       with:
-        go-version: 1.18
+        go-version: 1.21
       id: go
 
     - name: Check out code into the Go module directory

.gitignore

@@ -30,6 +30,7 @@ build
 # Generated files
 .docusaurus
 .cache-loader
+.yarn
 
 # Misc
 .env.local

.goreleaser.yml

@@ -114,11 +114,6 @@ nfpms:
       - dst: /var/log/dkron
         type: dir
 
-    overrides:
-      rpm:
-        replacements:
-          amd64: x86_64
-
 snapshot:
   name_template: "{{ .Tag }}-next"
 

CHANGELOG.md

@@ -6,6 +6,28 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## Unreleased
 
+## [3.2.7] - 2023-12-09
+
+### Changes
+
+- Replace dependent class library github.com/struCoder/pidusage with github.com/shirou/gopsutil/v3/process
+- pkg import only once https://github.com/distribworks/dkron/pull/1406
+- Bump deps
+- Use Go 1.21
+
+### Features
+
+- add TLS options for kafka executor plugin https://github.com/distribworks/dkron/pull/1384
+- Resource usage from child processes spawned by the task startup process.
+- Introduce two new metrics, dkron_job_execution_time_seconds and dkron_job_execution_done_count, for improved monitoring and alerting capabilities.
+
+### Fixes
+
+- cmd.Process.Kill() does not clean up the started process, causing the job to remain running. https://github.com/distribworks/dkron/pull/1385
+- The metric dkron_job_mem_usage_kb now exclusively calculates the memory usage of the task startup process, excluding 
+- Should output.TotalWritten be judged after cmd.Wait() is executed? https://github.com/distribworks/dkron/pull/1319
+- The metrics exposed by dkron are only of summary type and cannot cover related analysis for specific jobs. https://github.com/distribworks/dkron/pull/1310
+
 ## [3.2.6] - 2023-09-02
 
 ### Changes
@@ -22,11 +44,11 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Fixes
 
-Do not retry job when unknown status [#1335](https://github.com/distribworks/dkron/pull/1335)
+- Do not retry job when unknown status [#1335](https://github.com/distribworks/dkron/pull/1335)
 
 ### Features
 
-Prometheus metrics for shell executor [#1330](https://github.com/distribworks/dkron/pull/1330)
+- Prometheus metrics for shell executor [#1330](https://github.com/distribworks/dkron/pull/1330)
 
 ### Changes
 

Makefile

@@ -5,7 +5,7 @@ $(LINUX_PKGS):
 	fury push --as distribworks $@
 
 PACKAGE_NAME          := github.com/distribworks/dkron
-GOLANG_CROSS_VERSION  ?= v1.18.1
+GOLANG_CROSS_VERSION  ?= v1.21.5
 
 .PHONY: release-dry-run
 release-dry-run:
@@ -22,7 +22,7 @@ release-dry-run:
 		-v `pwd`:/go/src/$(PACKAGE_NAME) \
 		-w /go/src/$(PACKAGE_NAME) \
 		goreleaser/goreleaser-cross:${GOLANG_CROSS_VERSION} \
-		--rm-dist --skip-validate --skip-publish --timeout=1h
+		--rm-dist --skip-validate --skip-publish --timeout=1h --parallelism=1
 
 .PHONY: release
 release:

builtin/bins/dkron-executor-http/testdata/badssl.com-client.pem

@@ -1,13 +1,13 @@
 Bag Attributes
-    localKeyID: 77 23 D2 3A C0 2C 87 E2 AD 98 3F 06 68 F2 54 33 B6 05 0E FE 
+    localKeyID: 57 CC 99 3F 7C 3E D2 81 83 06 43 E2 9A 5E 12 06 9F 13 82 D0 
 subject=/C=US/ST=California/L=San Francisco/O=BadSSL/CN=BadSSL Client Certificate
 issuer=/C=US/ST=California/L=San Francisco/O=BadSSL/CN=BadSSL Client Root Certificate Authority
 -----BEGIN CERTIFICATE-----
-MIIEnTCCAoWgAwIBAgIJAPYAapdmy98xMA0GCSqGSIb3DQEBCwUAMH4xCzAJBgNV
+MIIEnTCCAoWgAwIBAgIJAMn8nqcEWekMMA0GCSqGSIb3DQEBCwUAMH4xCzAJBgNV
 BAYTAlVTMRMwEQYDVQQIDApDYWxpZm9ybmlhMRYwFAYDVQQHDA1TYW4gRnJhbmNp
 c2NvMQ8wDQYDVQQKDAZCYWRTU0wxMTAvBgNVBAMMKEJhZFNTTCBDbGllbnQgUm9v
-dCBDZXJ0aWZpY2F0ZSBBdXRob3JpdHkwHhcNMjExMjA0MDAwODE5WhcNMjMxMjA0
-MDAwODE5WjBvMQswCQYDVQQGEwJVUzETMBEGA1UECAwKQ2FsaWZvcm5pYTEWMBQG
+dCBDZXJ0aWZpY2F0ZSBBdXRob3JpdHkwHhcNMjMxMTI5MjIzNDAzWhcNMjUxMTI4
+MjIzNDAzWjBvMQswCQYDVQQGEwJVUzETMBEGA1UECAwKQ2FsaWZvcm5pYTEWMBQG
 A1UEBwwNU2FuIEZyYW5jaXNjbzEPMA0GA1UECgwGQmFkU1NMMSIwIAYDVQQDDBlC
 YWRTU0wgQ2xpZW50IENlcnRpZmljYXRlMIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8A
 MIIBCgKCAQEAxzdfEeseTs/rukjly6MSLHM+Rh0enA3Ai4Mj2sdl31x3SbPoen08
@@ -17,15 +17,48 @@ DeZTf2ES66abA5pOp60Q6OEdwg/vCUJfarhKDpi9tj3P6qToy9Y4DiBUhOct4MG8
 w5XwmKAC+Vfm8tb7tMiUoU0yvKKOcL6YXBXxB2kPcOYxYNobXavfVBEdwSrjQ7i/
 s3o6hkGQlm9F7JPEuVgbl/Jdwa64OYIqjQIDAQABoy0wKzAJBgNVHRMEAjAAMBEG
 CWCGSAGG+EIBAQQEAwIHgDALBgNVHQ8EBAMCBeAwDQYJKoZIhvcNAQELBQADggIB
-ABlLNovFvSrULgLvJmKX/boSWQOhWE0HDX6bVKyTs48gf7y3DXSOD+bHkBNHL0he
-m4HRFSarj+x389oiPEti5i12Ng9OLLHwSHK+7AfnrkhLHA8ML3NWw0GBr5DgdsIv
-7MJdGIrXPQwTN5j++ICyY588TfGHH8vU5qb5PrSqClLZSSHU05FTr/Dc1B8hKjjl
-d/FKOidLo1YDLFUjaB9x1mZPUic/C489lyPfWqPqoMRd5i/XShST5FPvfGuKRd5q
-XKDkrn+GaQ/4iDDdCgekDCCPhOwuulavNxBDjShwZt1TeUrZNSM3U4GeZfyrVBIu
-Tr+gBK4IkD9d/vP7sa2NQszF0wRQt3m1wvSWxPz91eH+MQU1dNPzg1hnQgKKIrUC
-NTab/CAmSQfKC1thR15sPg5bE0kwJd1AJ1AqTrYxI0VITUV8Gka3tSAp3aKZ2LBg
-gYHLI2Rv9jXe5Yx5Dckf3l+YSFp/3dSDkFOgEuZm2FfZl4vNBR+coohpB9+2jRWL
-K+4fIkCJba+Y2cEd5usJE18MTH9FU/JKDwzC+eO9SNLFUw3zGUsSwgZsBHP6kiQN
-suia9q4M5f+68kzM4+0NU8HwwyzZEtmTBhktKHijExixdvjlMAZ8hAOsFifsevI0
-02dUYvtxoHaeXh4jpYHVNnsIf/74uLagiPHtVf7+9UZV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 -----END CERTIFICATE-----
+Bag Attributes
+    localKeyID: 57 CC 99 3F 7C 3E D2 81 83 06 43 E2 9A 5E 12 06 9F 13 82 D0 
+Key Attributes: <No Attributes>
+-----BEGIN ENCRYPTED PRIVATE KEY-----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+-----END ENCRYPTED PRIVATE KEY-----

builtin/bins/dkron-executor-shell/shell.go

@@ -1,17 +1,10 @@
 package main
 
 import (
-	"encoding/base64"
 	"errors"
-	"fmt"
-	"log"
 	"os"
 	"os/exec"
 	"runtime"
-	"strconv"
-	"strings"
-	"syscall"
-	"time"
 
 	"github.com/armon/circbuf"
 	dkplugin "github.com/distribworks/dkron/v3/plugin"
@@ -53,111 +46,6 @@ func (s *Shell) Execute(args *dktypes.ExecuteRequest, cb dkplugin.StatusHelper)
 	return resp, nil
 }
 
-// ExecuteImpl do execute command
-func (s *Shell) ExecuteImpl(args *dktypes.ExecuteRequest, cb dkplugin.StatusHelper) ([]byte, error) {
-	output, _ := circbuf.NewBuffer(maxBufSize)
-
-	shell, err := strconv.ParseBool(args.Config["shell"])
-	if err != nil {
-		shell = false
-	}
-	command := args.Config["command"]
-	env := strings.Split(args.Config["env"], ",")
-	cwd := args.Config["cwd"]
-
-	executionInfo := strings.Split(fmt.Sprintf("ENV_JOB_NAME=%s", args.JobName), ",")
-	env = append(env, executionInfo...)
-
-	cmd, err := buildCmd(command, shell, env, cwd)
-	if err != nil {
-		return nil, err
-	}
-	err = setCmdAttr(cmd, args.Config)
-	if err != nil {
-		return nil, err
-	}
-	// use same buffer for both channels, for the full return at the end
-	cmd.Stderr = reportingWriter{buffer: output, cb: cb, isError: true}
-	cmd.Stdout = reportingWriter{buffer: output, cb: cb}
-
-	stdin, err := cmd.StdinPipe()
-	if err != nil {
-		return nil, err
-	}
-
-	defer stdin.Close()
-
-	payload, err := base64.StdEncoding.DecodeString(args.Config["payload"])
-	if err != nil {
-		return nil, err
-	}
-
-	stdin.Write(payload)
-	stdin.Close()
-
-	jobTimeout := args.Config["timeout"]
-	var jt time.Duration
-
-	if jobTimeout != "" {
-		jt, err = time.ParseDuration(jobTimeout)
-		if err != nil {
-			return nil, errors.New("shell: Error parsing job timeout")
-		}
-		cmd.SysProcAttr = &syscall.SysProcAttr{Setpgid: true}
-	}
-
-	log.Printf("shell: going to run %s", command)
-
-	err = cmd.Start()
-	if err != nil {
-		return nil, err
-	}
-
-	var jobTimeoutMessage string
-	var jobTimedOut bool
-
-	if jt != 0 {
-		slowTimer := time.AfterFunc(jt, func() {
-			// Kill child process to avoid cmd.Wait()
-			err := syscall.Kill(-cmd.Process.Pid, syscall.SIGKILL) // note the minus sign
-			if err != nil {
-				jobTimeoutMessage = fmt.Sprintf("shell: Job '%s' execution time exceeding defined timeout %v. SIGKILL returned error. Job may not have been killed", command, jt)
-			} else {
-				jobTimeoutMessage = fmt.Sprintf("shell: Job '%s' execution time exceeding defined timeout %v. Job was killed", command, jt)
-			}
-
-			jobTimedOut = true
-		})
-
-		defer slowTimer.Stop()
-	}
-
-	quit := make(chan int)
-
-	go CollectProcessMetrics(args.JobName, cmd.Process.Pid, quit)
-
-	err = cmd.Wait()
-	quit <- cmd.ProcessState.ExitCode()
-	close(quit) // exit metric refresh goroutine after job is finished
-
-	if jobTimedOut {
-		_, err := output.Write([]byte(jobTimeoutMessage))
-		if err != nil {
-			log.Printf("Error writing output on timeout event: %v", err)
-		}
-	}
-
-	// Warn if buffer is overwritten
-	if output.TotalWritten() > output.Size() {
-		log.Printf("shell: Script '%s' generated %d bytes of output, truncated to %d", command, output.TotalWritten(), output.Size())
-	}
-
-	// Always log output
-	log.Printf("shell: Command output %s", output)
-
-	return output.Bytes(), err
-}
-
 // Determine the shell invocation based on OS
 func buildCmd(command string, useShell bool, env []string, cwd string) (cmd *exec.Cmd, err error) {
 	var shell, flag string