tiup playground on low disk machine cannot schedule replicas
Nobody has claimed this yet.
- Dominant language
- Go
- Stars
- 466
- Forks
- 338
- Avg merge
- 3d 7h
- Merged PRs (30d)
- 8
Description
Bug Report
Please answer these questions before submitting your issue. Thanks!
- What did you do?
I'm using tiup playground on a machine with 466Gi total space and 34Gi available space. I created the cluster with 3 kv nodes:
tiup playground nightly --kv=3 --tiflash=0
- What did you expect to see?
With 3 kv nodes, and max-replicas = 3, I expected that any tables I create would have 3 replicas.
- What did you see instead?
Because PD detects low space on the tikv nodes, it only creates 1 out of the 3 expected replicas (see https://github.com/tikv/pd/issues/3444).
I think tiup should possibly set the pd option schedule.low-space-ratio = 0.99 or similar so that we have a better chance of the cluster operating in an environment with low available space.
- What version of TiUP are you using (
tiup --version)?
v1.3.2 tiup
Go Version: go1.13
Git Branch: release-1.3
GitHash: 2d88460
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 with the tiup playground command and the PD configuration it generates, then review how schedule.low-space-ratio is passed to PD. Reproduce the low-disk three-KV-node setup if possible and verify that the resulting cluster can schedule the expected replicas without changing behavior for normal environments.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- go
- Domain
- devops
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 42/100