PD:tso:ErrGenerateTimestamp errors should not appear on the application side
Nobody has claimed this yet.
- Dominant language
- Go
- Stars
- 362
- Forks
- 274
- Avg merge
- 4d 19m
- Merged PRs (30d)
- 12
Description
When upgrading a TiDB cluster, PD changes the leader. PD:tso:ErrGenerateTimestamp error will appear on application side.In fact, the PD leader switchover only took 1-2 s. Retries should be added to reduce the probability of this error occurring on application side.
[2023/08/10 04:03:48.262 +00:00] [ERROR] [client.go:857] ["[pd] getTS error"] [dc-location=global] [stream-addr=https://db-pd-2.db-pd-peer.tidbxxxxxxx.svc:2379] [error="[PD:client:ErrClientGetTSO]rpc error: code = Unknown desc = [PD:tso:ErrGenerateTimestamp]generate timestamp failed, requested pd is not leader of cluster: rpc error: code = Unknown desc = [PD:tso:ErrGenerateTimestamp]generate timestamp failed, requested pd is not leader of cluster"]
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 at client.go around line 857, where the logged getTS error is reported, and trace how PD timestamp requests handle a leader switch. Add retry behavior for the temporary ErrGenerateTimestamp failure, then verify that a 1–2 second PD leader switchover does not surface the error to the application.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- go
- Domain
- distributed-systems
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100