[P3][Logs] Replace per-dispatcher lifecycle logs with manager and changefeed summaries
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
Dispatcher lifecycle logs are currently emitted per object for events such as add, commit, ready, remove, drop, and handshake. When the number of dispatchers is high, the resulting logs are large in volume but low in marginal information. This task shifts lifecycle visibility to manager-level and changefeed-level summaries.
Modification points
- Replace per-dispatcher lifecycle narration with periodic or window-based summaries at the manager or changefeed layer.
- Keep counts for key lifecycle events such as
added,removed,reset,handshake, anddrop. - Reserve per-dispatcher details for abnormal cases or diagnostics mode.
Entry points
downstreamadapter/eventcollector/event_collector.go:259downstreamadapter/eventcollector/event_collector.go:287downstreamadapter/eventcollector/event_collector.go:302downstreamadapter/eventcollector/event_collector.go:329downstreamadapter/eventcollector/dispatcher_stat.go:603downstreamadapter/eventcollector/dispatcher_stat.go:621downstreamadapter/eventcollector/dispatcher_stat.go:659downstreamadapter/eventcollector/dispatcher_stat.go:670logservice/logpuller/region_event_handler.go:387
Done criteria
- Large dispatcher create and delete waves produce summary-oriented logs by default.
- Managers expose the important counts without logging each dispatcher lifecycle step.
- Detailed per-dispatcher logs remain available only for exceptions or targeted diagnostics.
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
Start by reading the listed entry points in downstreamadapter/eventcollector/event_collector.go and dispatcher_stat.go, then inspect logservice/logpuller/region_event_handler.go. Trace where dispatcher lifecycle events are logged and how manager or changefeed context is available. Done means large create/delete waves produce summary logs by default, with per-dispatcher details limited to exceptions or diagnostics.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- go
- Domain
- observability
- Issue type
- Refactor
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100