Implement handling strategy for retryable vs non-retryable exceptions in workerPartition#6270
Merged
san81 merged 5 commits intoJan 15, 2026
Conversation
vecheka
requested review from
KarstenSchnitter,
chenqi0805,
dinujoh,
dlvenable,
engechas,
graytaylor0,
kkondaka,
oeyh,
san81,
sb2k16 and
srikanthjg
as code owners
vecheka
force-pushed
the
feature/worker-partition-retry-strategy
branch
2 times, most recently
from
e91a9aa to
fc43488
Compare
…in workerPartition Signed-off-by: Vecheka Chhourn <vecheka@amazon.com>
vecheka
force-pushed
the
feature/worker-partition-retry-strategy
branch
from
fc43488 to
0930903
Compare
chrisale000
approved these changes
Dec 16, 2025
wjyao0316
reviewed
Dec 17, 2025
| * @param ex - exception thrown by workerScheduler | ||
| * @return boolean: true if we should fallback to backoffRetry | ||
| */ | ||
| private boolean delayRetry(Optional<EnhancedSourcePartition> sourcePartition, Exception ex) { |
Contributor
There was a problem hiding this comment.
nit.
Suggest to change the naming of this function and isWorkerPartitionLeaseExtended if we have a new revision.
delayRetry and backoffRetry can be renamed to something like shouldLocalRetry. It is because extending partition lease is also a sort of retry in partition level.
isWorkerPartitionLeaseExtended is not 100% true because there is scenario to give up the partition and lease not being extended. It should also be shouldLocalRetry
Contributor
Author
There was a problem hiding this comment.
That's a good point. I've renamed them for better readability. Thanks!
wjyao0316
approved these changes
Dec 17, 2025
…retry-strategy Signed-off-by: Vecheka Chhourn <vecheka@amazon.com>
vecheka
force-pushed
the
feature/worker-partition-retry-strategy
branch
from
60e5f63 to
ee15c10
Compare
san81
reviewed
Jan 13, 2026
| log.info("Updating workerPartition {}", workerPartition.getPartitionKey()); | ||
| Duration age = Duration.between(partitionCreationTime, Instant.now()); | ||
| if (age.compareTo(this.sourceConfig.getDurationToGiveUpRetry()) <= 0) { | ||
| log.info("Partition {} is within or equal to the configured max duration, scheduling retry", workerPartition.getPartitionKey()); |
Contributor
Author
There was a problem hiding this comment.
Makes sense, will update this.
san81
reviewed
Jan 13, 2026
Comment on lines
+144
to
+147
| if (progressState instanceof DimensionalTimeSliceWorkerProgressState) { | ||
| DimensionalTimeSliceWorkerProgressState workerProgressState = (DimensionalTimeSliceWorkerProgressState) progressState; | ||
| updateWorkerPartition(workerProgressState.getPartitionCreationTime(), workerPartition); | ||
| shouldLocalRetry = false; |
Collaborator
There was a problem hiding this comment.
Intention of the logic is to update the worker partition when partitionCreationTime exists in the state? or do this only for DimensionalTimeSliceWorkerProgressState ?
Contributor
Author
There was a problem hiding this comment.
Currently, only DimensionalTimeSliceWorkerProgressState has partitionCreationTime field but ideally we will followup with adding partitionCreationTime field to other SaasWorkerProgressState types.
…retry-strategy Signed-off-by: Vecheka Chhourn <vecheka@amazon.com>
vecheka
force-pushed
the
feature/worker-partition-retry-strategy
branch
from
3f4074e to
274b74b
Compare
wjyao0316
approved these changes
Jan 13, 2026
san81
approved these changes
Jan 15, 2026
san81
pushed a commit
to san81/data-prepper
that referenced
this pull request
Jan 27, 2026
…in workerPartition (opensearch-project#6270) Signed-off-by: Vecheka Chhourn <vecheka@amazon.com>
simonelbaz
pushed a commit
to simonelbaz/data-prepper
that referenced
this pull request
Jan 31, 2026
…in workerPartition (opensearch-project#6270) Signed-off-by: Vecheka Chhourn <vecheka@amazon.com> Signed-off-by: Simon ELBAZ <elbazsimon9@gmail.com>
simonelbaz
pushed a commit
to simonelbaz/data-prepper
that referenced
this pull request
Jan 31, 2026
…in workerPartition (opensearch-project#6270) Signed-off-by: Vecheka Chhourn <vecheka@amazon.com>
simonelbaz
pushed a commit
to simonelbaz/data-prepper
that referenced
this pull request
Jan 31, 2026
…in workerPartition (opensearch-project#6270) Signed-off-by: Vecheka Chhourn <vecheka@amazon.com>
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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.
Description
This is second part of the change to handle retryable vs non-retryable excpetions.
First part PR for more context: #6255
How
We are adding a new generic exception class
SaaSCrawlerExceptionto be shared by all connectors. This class is similar to previous API specific exception class (e.gOffice365Exception).CrawlerException will have two criterias:
We will utilize CrawlerException, and throw this up all the way to
WorkerSchedulewhere in the followup PR:sourceCoordinator.saveProgressStateForPartition(workerPartition, DURATION_TO_DELAY_RETRY). If it continues to fail up to 30 days (using partitionCreationTime field to confirm), we will give up the worker partition.Is this change backward compatible?
Yes. We ensure to keep the catch block on generic "Exception" so all other connectors will still use the default behaviour of backoff retry every 5ms for all exception types.
Testing
Unit tests, ran the below successfully:
Local testing:
Issues Resolved
N/A
Check List
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.