[GOBBLIN-1995]Kill the writer thread when timeout happens to release the lock #3871
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
added 9 commits
January 29, 2024 16:22
…x poll records during runtime
… unblock other trhead talking with HDFS
homatthew
reviewed
Jan 30, 2024
| state.setProp(CURRENT_PARTITIONED_WRITERS_COUNTER, partitionWriters.size() + 1); | ||
| return future.get(writeTimeoutInterval, TimeUnit.SECONDS); | ||
| } catch (ExecutionException | InterruptedException e) { | ||
| throw new RuntimeException("Error creating writer", e); | ||
| } catch (TimeoutException e) { | ||
| throw new RuntimeException(String.format("Failed to create writer due to timeout. The operation timed out after %s seconds.", writeTimeoutInterval), e); | ||
| } | ||
| finally { | ||
| if (future != null && !future.isDone()) { | ||
| future.cancel(true); |
There was a problem hiding this comment.
Does this have the intended effect? In the issues you were seeing, did we see threads that are sleeping / waiting on IO, which need to be interrupted via cancel?
I am asking this because if the thread needs to actively check if it's being requested to cancel (unless it's actively waiting / sleeping).
Read the below to see what I am describing
https://stackoverflow.com/questions/28043225/future-cancel-does-not-work
There was a problem hiding this comment.
@homatthew thanks for the information. The thread here is blocking on getting the HDFS mount table which is an IO operation. I also did a test to do a for loop to talk with HDFS and do timeout, and make sure cancel can work correctly in this case.
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Outdated suggestions cannot be applied.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.
Dear Gobblin maintainers,
Please accept this PR. I understand that it will not be reviewed until I have checked off all the steps below!
JIRA
Description
Now when there is timeout happens talking with HDFS, we timeout and fail the job but the thread won't be killed in this case, and all other calls to HDFS will be blocked by this hanging thread.
Tests
Commits