Call FlowExecutionListener.fireResumed a bit later
#260
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
dwnusbaum
reviewed
May 11, 2022
Comment on lines
+570
to
+572
|
|
||
| // we've been restarted while we were running. let's get the execution going again. | ||
| FlowExecutionListener.fireResumed(fetchedExecution); |
There was a problem hiding this comment.
Is this just so that the message gets printed first, or is it important for the AfterRestartTask to be in the queue before we start resuming the steps?
There was a problem hiding this comment.
Mainly just to print the message first.
dwnusbaum
approved these changes
May 11, 2022
|
I think this is OK to release at any time. |
jglick
added a commit
to jglick/workflow-durable-task-step-plugin
that referenced
this pull request
Jun 3, 2022
jglick
added a commit
to jenkinsci/workflow-durable-task-step-plugin
that referenced
this pull request
Oct 17, 2022
…tion of `node` step, not kill whole build (#180) * Sketch of non-`Pickle`-based resumption of `ExecutorStepExecution` * Amending test from #34 * Fixing `Callback` * Exploring where to detect a failure eligible for retry * Making `normalNodeDisappearance` pass under somewhat altered circumstances * First successful `node` block retry * `FilePathDynamicContext` must take precedence over `ExecutorStepDynamicContext.FilePathTranslator` * `ExecutorStepTest.unloadableExecutorPickle` as currently conceived no longer makes sense * Fixing `ExecutorStepTest.nodeDisconnectMissingContextVariableException` by copying some diagnostic code from `FilePathDynamicContext` * Fixing `ExecutorStepTest.contextualizeFreshFilePathAfterAgentReconnection` * Marking `ExecutorStepRetryEligibility` beta * Javadoc imports * SpotBugs: does not make sense to allow `ExecutorStepDynamicContext.node` to be null * Reasonable behavior of `ExecutorPickleTest.canceledQueueItem` * User-visible logging about 5m timeout now that `ExecutorPickle.printWaitingMessage` is unused * `ExecutorPickleTest` → `ExecutorStepDynamicContextTest` * Move `TIMEOUT_WAITING_FOR_NODE_MILLIS` from `ExecutorPickle` to `ExecutorStepExecution` * Some user-visible logging when `ExecutorStepRetryEligibility` is _not_ used * Do not block on `FutureImpl`. After jenkinsci/workflow-step-api-plugin#73 this caused `SecretsMasker` to block on a `KubernetesComputer` which in turn blocked provisioning of a `TaskListener`, causing `RestartPipelineTest.terminatedPodAfterRestart` to fail to find a message: ``` FINE o.j.p.w.s.d.DurableTaskStep$Execution#getWorkspace: rediscovering that terminated-pod-after-restart-1-8fb2m-j206k-8x125 has been removed and timeout has expired FINE o.j.p.w.s.d.DurableTaskStep$Execution#_listener: JENKINS-34021: could not get TaskListener in CpsStepContext[9:sh]:Owner[terminated Pod After Restart/1:terminated Pod After Restart #1] java.util.concurrent.TimeoutException at hudson.remoting.AsyncFutureImpl.get(AsyncFutureImpl.java:102) at org.jenkinsci.plugins.workflow.support.steps.ExecutorStepDynamicContext$Translator.get(ExecutorStepDynamicContext.java:115) Caused: java.io.IOException at org.jenkinsci.plugins.workflow.support.steps.ExecutorStepDynamicContext$Translator.get(ExecutorStepDynamicContext.java:118) at org.jenkinsci.plugins.workflow.steps.DynamicContext$Typed.get(DynamicContext.java:97) at org.jenkinsci.plugins.workflow.cps.ContextVariableSet.get(ContextVariableSet.java:139) at org.jenkinsci.plugins.workflow.cps.ContextVariableSet$1Delegate.doGet(ContextVariableSet.java:98) at org.jenkinsci.plugins.workflow.support.DefaultStepContext.get(DefaultStepContext.java:75) at org.csanchez.jenkins.plugins.kubernetes.pipeline.SecretsMasker$Factory.get(SecretsMasker.java:85) at org.csanchez.jenkins.plugins.kubernetes.pipeline.SecretsMasker$Factory.get(SecretsMasker.java:73) at org.jenkinsci.plugins.workflow.steps.DynamicContext$Typed.get(DynamicContext.java:95) at org.jenkinsci.plugins.workflow.cps.ContextVariableSet.get(ContextVariableSet.java:139) at org.jenkinsci.plugins.workflow.cps.CpsThread.getContextVariable(CpsThread.java:135) at org.jenkinsci.plugins.workflow.cps.CpsStepContext.doGet(CpsStepContext.java:297) at org.jenkinsci.plugins.workflow.support.DefaultStepContext.get(DefaultStepContext.java:75) at org.jenkinsci.plugins.workflow.support.DefaultStepContext.getListener(DefaultStepContext.java:127) at org.jenkinsci.plugins.workflow.support.DefaultStepContext.get(DefaultStepContext.java:87) at org.jenkinsci.plugins.workflow.steps.durable_task.DurableTaskStep$Execution._listener(DurableTaskStep.java:421) at org.jenkinsci.plugins.workflow.steps.durable_task.DurableTaskStep$Execution.listener(DurableTaskStep.java:412) at org.jenkinsci.plugins.workflow.steps.durable_task.DurableTaskStep$Execution.getWorkspace(DurableTaskStep.java:363) at org.jenkinsci.plugins.workflow.steps.durable_task.DurableTaskStep$Execution.check(DurableTaskStep.java:570) at org.jenkinsci.plugins.workflow.steps.durable_task.DurableTaskStep$Execution.run(DurableTaskStep.java:553) at … FINE o.j.p.w.s.d.DurableTaskStep$Execution$NewlineSafeTaskListener#getLogger: creating filtered stream FINE o.j.p.w.s.d.DurableTaskStep$Execution#_listener: terminated-pod-after-restart-1-8fb2m-j206k-8x125 has been removed for 15 sec, assuming it is not coming back ``` * A `sh` step could fail to find a workspace at the moment it starts. java.lang.NullPointerException at org.jenkinsci.plugins.workflow.steps.durable_task.DurableTaskStep$Execution.start(DurableTaskStep.java:329) at org.jenkinsci.plugins.workflow.cps.DSL.invokeStep(DSL.java:319) at org.jenkinsci.plugins.workflow.cps.DSL.invokeMethod(DSL.java:193) at org.jenkinsci.plugins.workflow.cps.CpsScript.invokeMethod(CpsScript.java:122) at … * Also handling `ClosedChannelException` as commonly thrown by `SimpleBuildStep`s and the like: jenkinsci/kubernetes-plugin#1083 (comment) * Also retry after `MissingContextVariableException` on `FilePath` * Remove `super.onResume` left over in #46; see jenkinsci/workflow-step-api-plugin#66 * Documenting need for JENKINS-30383 * jenkinsci/workflow-step-api-plugin#77 allows `retryNodeBlockSynchAcrossRestarts` to be fixed * jenkinsci/workflow-step-api-plugin#77 released * Expanding `MissingContextVariableException` type list after noticing that `isUnix` requests `Launcher` not `FilePath` * Never log a message with only `PlaceholderTask.cookie`; also need `.runId` for context * Suppressing uninteresting log message * Maybe need to call `StepContext.saveState`? * Deleting comment; for now it seems clearest to keep `ExecutorStepDynamicContext` as a distinct type * Resolve longstanding tech debt from #104 & #117 about `BodyExecution`. When we do not use pickles, there is no need for `transient`. Moving the reference to `ExecutorStepExecution` from `PlaceholderTask` avoids a memory leak. This is only possible now that `Callback` holds the `ExecutorStepExecution`. And we can now properly handle `AsynchronousExecution.interrupt`. * Sketch of updating `WorkspaceStepExecution`: #180 (comment) * Allowing `FilePathDynamicContext` to block waiting for an agent to reconnect. Not great since the CPS VM thread will only grant 5s. Better would be to avoid unnecessary use of `StepContext.get(FilePath.class)` (which implicitly requires a live `Channel`); perhaps `FilePathRepresentation` should become an API. Amends b65ff1c. #180 (comment) * Adding a comment about 8b08398 * 8b08398 seems to allow a968adf to be finished. * Refining 8b08398 to avoid tail call * `org.jenkinsci.plugins.workflow.support.concurrent.Futures` deprecated * SpotBugs reminds me that `WorkspaceStepExecution.Callback` needs an SVUID; taking one from trunk * Adding `NodeTranslator` * Beginning rewrite to `AgentErrorCondition` * Extracted `RetryExecutorStepTest` from `ExecutorStepTest` * Rely on blocking `ExecutorStepExecution.onResume` * Removing bogus `waitForMessage` (and unnecessary `interrupt`) from `unrestorableAgent`. Sometimes a failing `waitForMessage` seems to trigger https://maven.apache.org/surefire/maven-surefire-plugin/faq.html#corruptedstream though I cannot track down why. * `PlaceholderTask` may not hold a reference to `ExecutorStepExecution` * Equivalent of #184 for `ExecutorStepDynamicContext` * Assert jenkinsci/workflow-api-plugin@b1778a9 * jenkinsci/workflow-cps-plugin#534 & jenkinsci/workflow-job-plugin#260 released * Windows tests jenkinsci/workflow-basic-steps-plugin#203 (comment) * Avoiding `powershell` jenkinsci/workflow-basic-steps-plugin#203 (comment) * Better usage of Hamcrest; got a flake https://github.com/jenkinsci/workflow-durable-task-step-plugin/pull/180/checks?check_run_id=6779178701 with `QueueTaskCancelled` instead of `RemovedNodeCause` * In fact the flake was in a different test (`canceledQueueItem`) * Forgot to make `canceledQueueItem` not use `sh`, though it should not really matter since this step is expected to fail anyway * Test flake: https://github.com/jenkinsci/workflow-durable-task-step-plugin/pull/180/checks?check_run_id=7240329028 * Use `AgentOfflineException` from `ExecutorStepDynamicContext` as well * Can now run all of `AgentErrorConditionTest` * Responding to `CancellationException` in `ExecutorStepDynamicContext` with `QueueTaskCancelled`, analogous to handling in `ExecutorPickle` * Permit `canceledQueueItem` to pass on `RemovedNodeCause` * Normalizing indentation * jenkinsci/workflow-step-api-plugin#124 deprecates an overload which 2e95e3a only handles at one call site; TBD if others should be switched to `actualInterruption` * Refined `ExecutorStepDynamicContext` behavior in case of `quietingDown` #180 (comment) or `terminating` #180 (comment) * Add an explicit `serialVersionUID` to `ExecutorStepExecution.PlaceholderTask.Callback` #180 (comment) * Optimizing `withExecution` to only consider steps in the current build #180 (comment) * Pick up jenkinsci/workflow-api-plugin#256 #180 (comment) * jenkinsci/workflow-api-plugin#256 released
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Print the message about resuming the build, and schedule
AfterRestartTask, first. Complements jenkinsci/workflow-api-plugin#221.