[NO REVIEW] Fix 400/1024 error when container is re-created with same name#49131
Closed
xinlian12 wants to merge 3 commits intoAzure:mainfrom
Closed
[NO REVIEW] Fix 400/1024 error when container is re-created with same name#49131xinlian12 wants to merge 3 commits intoAzure:mainfrom
xinlian12 wants to merge 3 commits intoAzure:mainfrom
Conversation
Reset request context (resolvedCollectionRid, forceNameCacheRefresh, INTENDED_COLLECTION_RID_HEADER) in StaleResourceRetryPolicy before retry so the retry re-resolves the collection and sends the correct RID. Fixes Azure#49097 Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Contributor
There was a problem hiding this comment.
Pull request overview
This pull request addresses a Cosmos DB retry bug where requests can keep sending a stale x-ms-cosmos-intended-collection-rid header after a container is deleted and re-created with the same name (notably impacting gateway mode and 400/1024 retries). It resets request context state after refreshing the collection cache so the retry re-resolves the container RID and updates headers correctly.
Changes:
- Reset
RxDocumentServiceRequestretry state inStaleResourceRetryPolicy(force name cache refresh, clear resolved collection RID, remove intended-collection-rid header). - Add a parameterized unit test validating request-context/header reset behavior for both 410/1000 and 400/1024 scenarios.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.
| File | Description |
|---|---|
| sdk/cosmos/azure-cosmos/src/main/java/com/azure/cosmos/implementation/StaleResourceRetryPolicy.java | Resets request context and intended collection RID header after cache refresh to ensure retry re-resolves the container. |
| sdk/cosmos/azure-cosmos-tests/src/test/java/com/azure/cosmos/implementation/StaleResourceExceptionRetryPolicyTest.java | Adds parameterized unit coverage ensuring request context/header/session cleanup occurs on stale-container retry paths. |
| // and sends the updated intended-collection-rid header. | ||
| if (this.request != null) { | ||
| this.request.forceNameCacheRefresh = true; | ||
| this.request.requestContext.resolvedCollectionRid = null; |
xinlian12
commented
May 8, 2026
| // Reset request context so the retry re-resolves the collection | ||
| // and sends the updated intended-collection-rid header. | ||
| if (this.request != null) { | ||
| this.request.forceNameCacheRefresh = true; |
Member
Author
There was a problem hiding this comment.
do we still need this.request.forceNameCacheRefresh = true;? the collection cache is already refresh above
why not just set the this.request.requestContext.resolvedCollectionRid = refreshedCollectionrid?
…eRefresh - Set resolvedCollectionRid to refreshedCollectionRid instead of null - Remove unnecessary forceNameCacheRefresh (cache already refreshed) - Add requestContext null guard to prevent NPE Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Remove INTENDED_COLLECTION_RID_HEADER in RenameCollectionAwareClientRetryPolicy (404/1002 READ_SESSION_NOT_AVAILABLE) to prevent stale RID on retry. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.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
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.
Problem
When a container is deleted and re-created with the same name, subsequent read/write/query operations fail with:
(HTTP 400, sub-status 1024 / INCORRECT_CONTAINER_RID_SUB_STATUS)
This affects gateway mode because the stale intended-collection-rid header persists through retry.
Root Cause
StaleResourceRetryPolicycorrectly handles 400/1024 by refreshing the collection cache and clearing session tokens, but it did not reset the request context before the retry:request.requestContext.resolvedCollectionRidstill held the old collection RIDrequest.forceNameCacheRefreshwas stillfalsex-ms-cosmos-intended-collection-ridHTTP header still carried the old RIDOn retry,
RxGatewayStoreModel.addIntendedCollectionRid()checks if the header is already set and skips updating it — so the stale RID is sent again, causing the same 400/1024 error.Fix
After refreshing the collection cache in
StaleResourceRetryPolicy.shouldRetry(), reset:resolvedCollectionRid→nullforceNameCacheRefresh→trueINTENDED_COLLECTION_RID_HEADERThis ensures the retry re-resolves the collection and sends the correct (new) RID.
Testing
Added parameterized unit test
requestContextResetOnRetrycovering both error codes:The test verifies that after
shouldRetry():resolvedCollectionRidis cleared tonullforceNameCacheRefreshis set totrueINTENDED_COLLECTION_RID_HEADERis removedFixes #49097