Files
my-rook-config/pkg/operator/ceph/object
Sébastien Han e478bdadb1 ceph: stop the monitoring before cleanup
Saw this today in the logs:

```
2021-06-02 13:02:54.787838 I | ceph-block-pool-controller: deleting pool "testpool"
2021-06-02 13:02:56.178094 I | cephclient: no images/snapshosts present in pool "testpool"
2021-06-02 13:02:56.178125 I | cephclient: purging pool "testpool" (id=17)
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x0 pc=0x1cef49f]

goroutine 1607 [running]:
github.com/rook/rook/pkg/operator/ceph/pool.(*mirrorChecker).checkMirroringHealth(0xc001a98ba0, 0x8, 0xc000d34c60)
	/home/runner/work/rook/rook/pkg/operator/ceph/pool/health.go:115 +0xdf
github.com/rook/rook/pkg/operator/ceph/pool.(*mirrorChecker).checkMirroring(0xc001a98ba0, 0xc0016fd500)
	/home/runner/work/rook/rook/pkg/operator/ceph/pool/health.go:82 +0x147
created by github.com/rook/rook/pkg/operator/ceph/pool.(*ReconcileCephBlockPool).reconcile
	/home/runner/work/rook/rook/pkg/operator/ceph/pool/controller.go:298 +0xee5
```

Essentially, it's intermittent but when deleting the pool the
healthcheck kicked in and fetched the mirroring status, which returned
empty. THe subsequent code tried to access content of a nil pointer,
  hence the error.
So now we stop monitoring first, then we proceed with the deletion.

Signed-off-by: Sébastien Han <seb@redhat.com>
2021-06-02 16:14:10 +02:00
..
2021-05-27 04:46:13 +00:00
2021-05-27 04:46:13 +00:00
2021-03-31 10:58:52 +02:00
2020-11-06 10:01:04 +01:00