Created attachment 1801897 [details] default_pod_created_without_annotations Description of problem (please be detailed as possible and provide log snippests): In a Multus enabled OCS Internal mode cluster, all commands are not working as expected in toolbox pod. Simple ceph commands works, but rbd commands get hung. ``` sh-4.4# ceph -s cluster: id: 465f536e-3fed-41ec-88c5-00584ad4f069 health: HEALTH_OK services: mon: 3 daemons, quorum a,b,c (age 6h) mgr: a(active, since 6h) mds: ocs-storagecluster-cephfilesystem:1 {0=ocs-storagecluster-cephfilesystem-a=up:active} 1 up:standby-replay osd: 3 osds: 3 up (since 6h), 3 in (since 6h) rgw: 1 daemon active (ocs.storagecluster.cephobjectstore.a) data: pools: 10 pools, 176 pgs objects: 351 objects, 204 MiB usage: 3.5 GiB used, 1.5 TiB / 1.5 TiB avail pgs: 176 active+clean io: client: 852 B/s rd, 11 KiB/s wr, 1 op/s rd, 1 op/s wr sh-4.4# rbd ls -p ocs-storagecluster-cephblockpool >> hangs & no output after several minutes ``` It was observed that toolbox pod doesn't have Multus related annotaions/interfaces by default. Operator should take care of this automatically and apply proper multus annotations to the toolbox. Version of all relevant components (if applicable): OCP: 4.8.0-0.nightly-2021-07-14-153019 OCS: ocs-operator.v4.8.0-452.ci Does this issue impact your ability to continue to work with the product (please explain in detail what is the user impact)? User won't be able to execute all commands via toolbox Is there any workaround available to the best of your knowledge? Yes, Add annotation to "rook-ceph-tools" deployment similar to what MONs have and set hostNetwork to false. Rate from 1 - 5 the complexity of the scenario you performed that caused this bug (1 - very simple, 5 - very complex)? 1 Can this issue reproducible? Yes Can this issue reproduce from the UI? Yes If this is a regression, please provide more details to justify this: - Steps to Reproduce: 1. Install OCS operator 2. Create Storage Cluster with Multus 3. Use this command to start toolbox pod $ oc patch ocsinitialization ocsinit -n openshift-storage --type json --patch '[{ "op": "replace", "path": "/spec/enableCephTools", "value": true }]' 4. Run a simple rbd command $ rbd ls -p ocs-storagecluster-cephblockpool Actual results: rbd command get hung in toolbox pod Expected results: all ceph & rbd commands should work same with and without multus Additional info:
Rook does not create the toolbox, ocs-operator does, on-demand. So this request will have to be done from the ocs-operator.
Not a 4.8 blocker, can be fixed in 4.8.1
This is definitely something we should fix. But, since multus is only going GA in ODF 4.10, moving it accordingly since we have time to fix it there. Also giving devel_ack+. Rohan, if you could take a quick look, see if you can take care of this and set it to ASSIGNED. Otherwise let me know and I'll see if I can find anyone else.
This BZ is directly related to the Multus Epic, which is not targeted for ODF 4.10: https://issues.redhat.com/browse/RHSTOR-2067 As we have hit feature freeze for 4.10, moving this to ODF 4.11.
Rohan, any plan to work on this? Thanks
Ideally yes if someone from ocs-op can pick it up, it looks small. It's only about adding the network annotations to the toolbox when requested. José, do we have someone else than Rohan? He moved to a different team so I'm not sure if we can count on him :) Thanks!
Reproducer looks clear.
Eran and I talked about targeting 4.13.z for this. @uchapaga could you help me be sure this is in the work queue for a 4.13 z-stream?
Since we have all the acks on this BZ, will take this up for 4.14. I will create a 4.13.z clone of this BZ when it's available.