[FLINK-38867] [Kubernetes Operator] Fix restartSavepointNonce to respect initialSavepointPath in BlueGreenDeployments#12
Open
james-kan-shopify wants to merge 1 commit into
Conversation
drossos
reviewed
Jan 12, 2026
drossos
left a comment
drossos
left a comment
There was a problem hiding this comment.
Left a larger comment about how we could refactor this to potentially streamline the logic
Really like direction 👍
james-kan-shopify
force-pushed
the
flink-blue-green-restartsavepointnonce
branch
from
84834f0 to
4d3cc92
Compare
james-kan-shopify
changed the title
james-kan-shopify
changed the title
drossos
reviewed
Jan 14, 2026
drossos
left a comment
drossos
left a comment
There was a problem hiding this comment.
Comment on gap with redeploying from stateless
| FlinkBlueGreenDeploymentSpec spec2 = createBasicSpec(); | ||
|
|
||
| // Change both top-level (configuration) and nested spec | ||
| // With new logic, only nested spec changes matter - setSavepointRedeployNonce triggers |
There was a problem hiding this comment.
This isn't specific to this code, but in our PR / JIRA in OSS we should highlight that for savepoint upgrade behaviour is not working as expected (loads from taken savepoint in transition and not specified one by initalSavepointPath)
drossos
force-pushed
the
flink-blue-green-restartsavepointnonce
branch
from
0e4b463 to
5e48bdb
Compare
Comment on lines
+352
to
+355
| // lastCheckpoint is null in two scenarios: | ||
| // 1. savepointRedeployNonce changed - user wants to redeploy from initialSavepointPath | ||
| // 2. upgradeMode is STATELESS - don't take savepoints (but use initialSavepointPath if | ||
| // set) |
Author
There was a problem hiding this comment.
@drossos we gotta revise this comment, likely drop the comment in the else case and move it up here to centralize it.
…estartSavepointNonce Co-authored-by: Daniel Rossos <daniel.rossos@shopify.com>
james-kan-shopify
force-pushed
the
flink-blue-green-restartsavepointnonce
branch
from
5e48bdb to
ed3508a
Compare
drossos
approved these changes
Jan 23, 2026
drossos
left a comment
drossos
left a comment
There was a problem hiding this comment.
LGTM (although co-authored a small part), good to deploy on our side
Draft
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
2 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.
What is the purpose of the change
This pull request fixes a bug where
FlinkBlueGreenDeploymentignores the user-specifiedinitialSavepointPathwhensavepointRedeployNonceis incremented. The controller now properly honorssavepointRedeployNoncechanges and uses the specified savepoint location for redeployment, aligning Blue/Green deployment behavior with standardFlinkDeploymentbehavior.Brief change log
BlueGreenDiffType.SAVEPOINT_REDEPLOYto handle nonce-triggered redeploymentsFlinkBlueGreenDeploymentSpecDiffto detect and prioritizesavepointRedeployNoncechangesstartSavepointRedeployTransition()to skip savepoint triggering and useinitialSavepointPathfrom specVerifying this change
This change added tests and can be verified as follows:
verifySavepointRedeployNonceTriggersTransitionWithInitialSavepointPathintegration test inFlinkBlueGreenDeploymentControllerTestthat verifies the full savepoint redeploy flow:ACTIVE_BLUEstatesavepointRedeployNoncewith user-specifiedinitialSavepointPathTRANSITIONING_TO_GREENinitialSavepointPathACTIVE_GREENDoes this pull request potentially affect one of the following parts:
CustomResourceDescriptors: noDocumentation