Skip to content

fix: omnisharded resharding copy and replication#869

Merged
levkk merged 4 commits intomainfrom
levkk-omni-upserts
Apr 6, 2026
Merged

fix: omnisharded resharding copy and replication#869
levkk merged 4 commits intomainfrom
levkk-omni-upserts

Conversation

@levkk
Copy link
Copy Markdown
Collaborator

@levkk levkk commented Apr 5, 2026 •
edited
Loading

  • fix: bubble up resharding COPY error; we were swallowing it previously
  • fix: omnisharded table copies are distributed between shards evenly, avoiding duplicate copies of the same table data
  • fix: omnisharded table replication uses upserts to 1) avoid unique key violation, 2) prevent race conditions between inserts into omni and sharded tables

@codecov
Copy link
Copy Markdown

codecov bot commented Apr 5, 2026 •
edited
Loading

Codecov Report

❌ Patch coverage is 30.66667% with 52 lines in your changes missing coverage. Please review.

Files with missing lines Patch % Lines
...nd/replication/logical/publisher/publisher_impl.rs 0.00% 30 Missing ⚠️
...end/replication/logical/publisher/parallel_sync.rs 0.00% 15 Missing ⚠️
...src/backend/replication/logical/subscriber/copy.rs 0.00% 4 Missing ⚠️
...src/backend/replication/logical/publisher/table.rs 80.00% 3 Missing ⚠️

📢 Thoughts on this report? Let us know!

@levkk levkk changed the title fix: distribute omni tables during resharding multiple resharding fixes Apr 6, 2026
@levkk levkk changed the title multiple resharding fixes fix: omnisharded copy Apr 6, 2026
@levkk levkk changed the title fix: omnisharded copy fix: omnisharded resharding copy Apr 6, 2026
@levkk levkk changed the title fix: omnisharded resharding copy fix: omnisharded resharding copy and replication Apr 6, 2026
@levkk levkk merged commit bf25cbd into main Apr 6, 2026
8 checks passed
@levkk levkk deleted the levkk-omni-upserts branch April 6, 2026 01:14
@levkk levkk mentioned this pull request Apr 7, 2026
Merged
levkk added a commit that referenced this pull request Apr 7, 2026
@levkk
fix: table LSN not preserved during copy -> replication handover (#874)
004f03b 
- fix: regression introduced in #869 erasing table LSNs during copy to
replication handover, causing rows written between slot creation and
copy start replayed on the destination
- fix: off-by-one in replication to copy LSN check
- fix: advancing LSN prior to transaction commit, potentially skipping
rows in the same transaction
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@levkk