Skip to content

v2.5.4 (Worker Batching)

Choose a tag to compare

View all tags
@chihebnabil chihebnabil released this 27 Jun 23:22
· 59 commits to main since this release
v2.5.4
9dc0c18

Breaking Changes

  • Worker mode now uses batching instead of immediate sending
  • Logs and spans are queued and flushed at request end
  • Default batch size: 10 (workers) vs 100 (non-workers)
  • No timer-based flush in workers (flush at request end via ctx.waitUntil)

Improvements

  • 78% fewer HTTP requests in workers (1 batch vs 1 per log)
  • Better reliability - logs queued until flush
  • Aligns with industry standard (use batching in workers)

Configuration

No changes needed. Existing configs work automatically:

  • workerMode: true → uses batching with size 10
  • workerMode: false → uses batching with size 100

API Changes

None. All existing methods work the same:

  • logger.info(), logger.error(), etc.
  • workerFetch(), pagesFunction()
  • logger.flush() - now more important in workers

Bug Fixes

  • Fixed issue where logs could be lost in workers due to isolate termination
  • Fixed race condition in Simple processors

Full Changelog: v2.5.3...v2.5.4

Assets 2
Loading