[P4][Logs] Add periodic changefeed-level summaries for distributed state and progress
Nobody has claimed this yet.
- Dominant language
- Go
- Stars
- 56
- Forks
- 63
- Avg merge
- 2d 20h
- Merged PRs (30d)
- 34
Description
Parent issue: #4691
Intent
Today, changefeed state is scattered across maintainer, dispatcher manager, event collector, and controller logs. That forces operators to stitch together many lines to understand a single changefeed. This task adds periodic changefeed-level summaries so the core state is visible in one place.
Modification points
- Add periodic summaries that expose the main distributed-state counters for a changefeed.
- Include counts such as
dispatcherCount,tableCount,operatorCount,pendingBlockCount,resetCount, anderrorCount. - Use the summary path as the default operational view, keeping detailed component logs for anomalies and deeper diagnosis.
Entry points
maintainer/maintainer.go:268maintainer/maintainer.go:388maintainer/maintainer.go:404maintainer/maintainer.go:935maintainer/replica/region_count_refresher.go:122coordinator/controller.go:171
Done criteria
- A single periodic summary provides the main operational picture for a changefeed.
- Operators no longer need to combine many dispersed logs to answer basic state and progress questions.
- Detailed component logs remain available for drill-down when the summary indicates a problem.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Read the listed entry points in maintainer/maintainer.go, maintainer/replica/region_count_refresher.go, and coordinator/controller.go to trace how changefeed state is currently logged and refreshed. Run the relevant existing tests or development checks before changing the logging path. Done means one periodic summary exposes the named counters while detailed component logs remain available for diagnosis.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- go
- Domain
- backend, distributed-systems, observability
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 52/100