etcd-io / etcd-io/etcd-operator
Support handling data corruption case
- Dominant language
- Go
- Stars
- 164
- Forks
- 72
- Avg merge
- 22h 46m
- Merged PRs (30d)
- 34
Description
We depend on the CORRUPTION alarm to check whether there is data corruption.
[The steps](https://github.com/etcd-io/etcd-operator/blob/c8a968fc5575ff84bba294f0ea9d3ec5d1ffd409/internal/controller/etcdcluster_controller.go#L395-L398):
- first remove the member from the etcd cluster first
- Change the EtcdMember’s phase to Replacing (we reuse the Replacing workflow to take care of the remaining work)
- Disalarm the CORRUPTION alarm
Contributor guide
Research direction
Start with internal/controller/etcdcluster_controller.go at lines 395-398 and trace how the CORRUPTION alarm is handled. Verify the member-removal and Replacing workflow before implementing support for this case; done means the CORRUPTION alarm is disarmed after the member enters the replacement flow.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- go
- Domain
- distributed-systems
- Issue type
- Feature
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Active
- Clarity
- Mostly clear
- Newbie friendliness
- 68/100