Accept object-valued value and overheadFields in kafka entries - #2807
Open
delthas wants to merge 2 commits into
Open
Accept object-valued value and overheadFields in kafka entries#2807delthas wants to merge 2 commits into
delthas wants to merge 2 commits into
Conversation
createFromKafkaEntry coverage lived under tests/unit/replication/, while tests/unit/lib/models/ mirrors the source layout and already holds the sibling model specs. Pure relocation: only the two relative require paths change. Issue: BB-847
createFromKafkaEntry parsed the kafka message body and then parsed the nested value (and overheadFields) a second time, requiring producers to send them as JSON text. The D/R metadata stream is built by a MongoDB change-stream pipeline, which cannot serialise a document to a string — there is no $toJson and $convert rejects documents — so it can only emit these fields as objects. Decode them only when they are strings and pass objects through. This is a superset of the existing format: producers that stringify, S3C ingestion and the oplog reader, take the identical path. A field that is neither a string nor an object now yields the usual malformed-entry error naming the offending type. Previously JSON.parse coerced it, so an entry whose value was a number reached ObjectQueueEntry and passed checkSanity, since only bucket and key are checked. Issue: BB-847
Contributor
Hello delthas,My role is to assist you with the merge of this Available options
Available commands
Status report is not available. |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files
... and 2 files with indirect coverage changes
@@ Coverage Diff @@
## development/9.5 #2807 +/- ##
===================================================
- Coverage 75.76% 75.75% -0.02%
===================================================
Files 200 200
Lines 13922 13927 +5
===================================================
+ Hits 10548 10550 +2
- Misses 3364 3367 +3
Partials 10 10
Flags with carried forward coverage won't be shown. Click here to find out more. 🚀 New features to boost your workflow:
|
Contributor
Waiting for approvalThe following approvals are needed before I can proceed with the merge:
|
francoisferrand
approved these changes
Aug 19, 2026
francoisferrand
left a comment
francoisferrand
left a comment
Contributor
There was a problem hiding this comment.
note: should land on 9.6 branch (waiting for "flaky" fix and 9.5.0 release) to create it
Contributor
Author
Done |
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
3 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.
createFromKafkaEntryparses the kafka message body and then parses the nestedvalue— and, fordelentries,overheadFields— a second time, so producers have to send them as JSON text.The D/R metadata stream is built by a MongoDB change-stream pipeline (ZKOP-562), which cannot serialise a document to a string: there is no
$toJson, and$convertrejects documents. Manufacturing one for the consumer to parse straight back would mean either$function(per-document server-side JS on a production secondary, inside a change-stream pipeline) or a custom Kafka Connect SMT shipped in the kafka-connect image.The string requirement only exists because every producer so far builds entries in JavaScript —
ListRecordStreamdoesvalue: JSON.stringify(objectMd)on the oplog path, and S3C ingestion stringifies upstream. Nothing consumes the string form; it is parsed on receipt.So decode these fields only when they are strings, and pass objects through. That is a superset of the existing format: producers that stringify take the identical path, so ingestion is unaffected.
The first commit moves the
QueueEntryspec totests/unit/lib/models/, next to the model it tests.Issue: BB-847