force-leaving an active node
- Dominant language
- Go
- Stars
- 6.1k
- Forks
- 609
- Avg merge
- 15h 1m
- Merged PRs (30d)
- 3
Description
I was playing with `force-leave`, and I noticed that force-leaving an `active` node makes it stay in a permanent `leaving` state. Although I understand that `force-leave` is to be used with `failed` nodes, I can't find a reason why it couldn't work with `active` nodes too.
I am running two serf nodes on the same machine, and joined them in a cluster:
``` bash
$ serf agent -bind=127.0.0.1:8000 -rpc-addr=127.0.0.1:8001 -node=node1
$ serf agent -bind=127.0.0.1:9000 -rpc-addr=127.0.0.1:9001 -node=node2
```
Then I force-leave the active `node2` node:
``` bash
$ serf force-leave -rpc-addr=127.0.0.1:8001 node2
```
And this is the members list:
``` bash
$ serf members -rpc-addr=127.0.0.1:8001
node1 127.0.0.1:8000 alive
node2 127.0.0.1:9000 leaving
```
`node2` seems to stay in the `leaving` state forever, and never actually goes into `left` state.
Contributor guide
Assessment
This issue has not been assessed yet.