Skip to content

PLEX-2791 N-06 Improve LogPoller Heath Observability#436

Open
dhaidashenko wants to merge 1 commit intofix/plex-2783-l-05-fix-missed-finality-violation-in-case-of-empty-blockfrom
fix/PLEX-2791-N-06-improve-lp-health-observability
Open

PLEX-2791 N-06 Improve LogPoller Heath Observability#436
dhaidashenko wants to merge 1 commit intofix/plex-2783-l-05-fix-missed-finality-violation-in-case-of-empty-blockfrom
fix/PLEX-2791-N-06-improve-lp-health-observability

Conversation

@dhaidashenko
Copy link
Copy Markdown
Contributor

Problem

batchInsert Retry Path Can Silently Repeat Failed Persistence

The ORM batchInsert helper only shrinks the batch size when the returned error matches context.DeadlineExceeded and never below 500 rows. If a large insert fails for a size-sensitive database reason that does not surface as DeadlineExceeded, batchInsert returns the error immediately even though smaller sub-batches might succeed. Since the pollAndSaveLog function treats insert failures as retry-later and the main loop resumes from SelectLatestBlock+1, the poller can repeatedly recreate the same failing insert shape and stall ingestion progress.

Fix

Expose a general metric that signals if LogPoller is stuck.

@github-actions
Copy link
Copy Markdown
Contributor

✅ API Diff Results - No breaking changes


📄 View full apidiff report

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant