ceph: retry object health check if creation fails #8708
Conversation
|
This pull request has merge conflicts that must be resolved before it can be merged. @BlaineEXE please rebase it. https://rook.io/docs/rook/master/development-flow.html#updating-your-fork |
BlaineEXE
force-pushed
the
health-checker-failure-should-fail-reconcile
branch
2 times, most recently
from
6f9eb5e
to
7b4f2e6
Compare
BlaineEXE
changed the title
|
done |
BlaineEXE
force-pushed
the
health-checker-failure-should-fail-reconcile
branch
from
7b4f2e6
to
7e8cfcf
Compare
If the CephObjectStore health checker fails to be created, return a reconcile failure so that the reconcile will be run again and Rook will retry creating the health checker. This also means that Rook will not list the CephObjectStore as ready if the health checker can't be started. Signed-off-by: Blaine Gardner <blaine.gardner@redhat.com>
BlaineEXE
force-pushed
the
health-checker-failure-should-fail-reconcile
branch
from
7e8cfcf
to
5383ba2
Compare
| TypeMeta: controllerTypeMeta, | ||
| } | ||
| objectStore.Spec.Gateway.Port = 80 | ||
| setupNewEnvironment := func(additionalObjects ...runtime.Object) *ReconcileCephObjectStore { |
There was a problem hiding this comment.
Add this function to setup a new test environment ReconcileCephObjectStore so each test is independent.
| res, err := r.Reconcile(ctx, req) | ||
| assert.NoError(t, err) | ||
| assert.True(t, res.Requeue) | ||
| }) | ||
|
|
||
| t.Run("success - object store is running", func(t *testing.T) { | ||
| // set up an environment that has a ready ceph cluster, and return the reconciler for it | ||
| setupEnvironmentWithReadyCephCluster := func() *ReconcileCephObjectStore { |
There was a problem hiding this comment.
Add this additional function to help set up test environment ReconcileCephObjectStores where the CephCluster should be ready, allowing reconcile to proceed. (used in the previously existing successful test and the new test added)
BlaineEXE
changed the title
There was a problem hiding this comment.
LGTM. As a follow-up, we need to take the same approach for all health checker, ceph status, osd status, pool mirror, fs mirror. @BlaineEXE are you planning this?
I didn't notice any place where this would be a quick fix for any of the other health checkers, I think because RGW relies on an HTTP API where the rest use the ceph cli IIUC. |
|
I do want to backport this to 1.7 I think. |
ceph: retry object health check if creation fails (backport #8708)
If the CephObjectStore health checker fails to be created, return a
reconcile failure so that the reconcile will be run again and Rook will
retry creating the health checker. This also means that Rook will not
list the CephObjectStore as ready if the health checker can't be
started.
Signed-off-by: Blaine Gardner blaine.gardner@redhat.com
Description of your changes:
Which issue is resolved by this Pull Request:
Resolves #
Checklist:
make codegen) has been run to update object specifications, if necessary.