balancer-leader should be disabled when upgrade tikv instance trigger evict-leader
Open
Nobody has claimed this yet.
- Dominant language
- Go
- Stars
- 466
- Forks
- 338
- Avg merge
- 3d 7h
- Merged PRs (30d)
- 8
Description
When upgrade a TiDB cluster with multiple TiKV instances, after the first TiKV instance upgrade, region unbalance among all instances, so the balance-leader start to work. This is not necessary, since each TiKV waiting to be upgraded will evict-leader again.
So we should disable it, to make upgrade TiKV instance more stable and more fast.
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 tracing TiKV instance upgrade handling in TiUP, focusing on the interaction between balance-leader and evict-leader. Reproduce a multi-TiKV upgrade and verify that balance-leader stays disabled while instances are being upgraded and evict-leader is triggered, without destabilizing the upgrade.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- go
- Domain
- distributed-systems, infrastructure
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 30/100