Back to bug 1978722

Who When What Removed Added
Travis Nielsen 2021-07-02 16:50:21 UTC Assignee tnielsen shan
Sébastien Han 2021-07-05 13:42:10 UTC Flags needinfo?(sagrawal)
Sidhant Agrawal 2021-07-05 14:09:03 UTC Flags needinfo?(sagrawal)
Sidhant Agrawal 2021-07-05 14:40:36 UTC Keywords TestBlocker
Mudit Agarwal 2021-07-05 14:56:02 UTC CC muagarwa, shan
Flags needinfo?(shan)
Sébastien Han 2021-07-05 15:34:28 UTC Status NEW ASSIGNED
Flags needinfo?(shan)
Neha Berry 2021-07-06 10:32:49 UTC CC nberry
QA Contact ebenahar sagrawal
Raz Tamir 2021-07-06 14:15:05 UTC CC ratamir
RHEL Program Management 2021-07-06 14:15:14 UTC Target Release --- OCS 4.8.0
Sébastien Han 2021-07-06 21:29:44 UTC Status ASSIGNED POST
Link ID Github rook/rook/pull/8272
Red Hat Bugzilla 2021-07-07 02:10:55 UTC Doc Type If docs needed, set a value No Doc Update
Sébastien Han 2021-07-07 10:19:29 UTC Doc Text When the CephCluster is configured with Multus and multiple networks are
used to deploy Ceph some commands are failing to be executed from the
Operator. These commands, in particular, radosgw-admin ones need access
to the "ceph public network" to talk to OSDs. Unfortunately, the
Rook-Ceph Operator does not have the network annotations and thus doesn't
have the networks available and cannot reach OSDs. So the commands end
up hanging and eventually time out.
Applying the annotations to the Operator pod is possible but will result
in restarting the operator too and this should be avoided at all costs.
Also, applying the annotations beforehand is not possible since the
Multus declaration is in the CephCluster specification. So we would have
no idea what to do.

So the current approach runs a new sidecar container in the mgr pod to
act as a proxy for "some" ceph commands, only the radosgw-admin ones
for multi-site setup. This is a small container with admin access
running idle waiting for commands to be executed. In a sense, it is
similar to the toolbox but we didn't want to clearly expose it, so
running as a sidecar is quite nice.

Proxying command is obviously not always recommended since we add an
extra hop in the network path. Now each request has to go from the
operator pod to the API server to the remote pod to Ceph. Previously,
the command only goes from the operator to Ceph.

It's worth noting that external mode is not impacted since no rgw pod
is configured. This scenario is flexible and allows us to scale
pretty well since any CephCluster with Multus will see its mgr sidecar
deployed and can then talk to Ceph. We are not limited.
Mudit Agarwal 2021-07-07 10:51:52 UTC Doc Type No Doc Update Bug Fix
Mudit Agarwal 2021-07-07 15:09:17 UTC Doc Text When the CephCluster is configured with Multus and multiple networks are
used to deploy Ceph some commands are failing to be executed from the
Operator. These commands, in particular, radosgw-admin ones need access
to the "ceph public network" to talk to OSDs. Unfortunately, the
Rook-Ceph Operator does not have the network annotations and thus doesn't
have the networks available and cannot reach OSDs. So the commands end
up hanging and eventually time out.
Applying the annotations to the Operator pod is possible but will result
in restarting the operator too and this should be avoided at all costs.
Also, applying the annotations beforehand is not possible since the
Multus declaration is in the CephCluster specification. So we would have
no idea what to do.

So the current approach runs a new sidecar container in the mgr pod to
act as a proxy for "some" ceph commands, only the radosgw-admin ones
for multi-site setup. This is a small container with admin access
running idle waiting for commands to be executed. In a sense, it is
similar to the toolbox but we didn't want to clearly expose it, so
running as a sidecar is quite nice.

Proxying command is obviously not always recommended since we add an
extra hop in the network path. Now each request has to go from the
operator pod to the API server to the remote pod to Ceph. Previously,
the command only goes from the operator to Ceph.

It's worth noting that external mode is not impacted since no rgw pod
is configured. This scenario is flexible and allows us to scale
pretty well since any CephCluster with Multus will see its mgr sidecar
deployed and can then talk to Ceph. We are not limited.
Doc Type Bug Fix No Doc Update
Red Hat Bugzilla 2021-07-07 15:09:17 UTC Doc Type No Doc Update No Doc Update
OpenShift BugZilla Robot 2021-07-07 19:57:36 UTC Status POST MODIFIED
Sébastien Han 2021-07-07 19:58:03 UTC Link ID Github openshift/rook/pull/274
Orit Wasserman 2021-07-08 14:14:00 UTC CC owasserm
Sébastien Han 2021-07-08 15:54:34 UTC Summary OCS deployment with multus unsuccessful with noobaa stuck in Configuring phase CephObjectStore deployment fails with multus, leading Noobaa to never be ready
Mudit Agarwal 2021-07-09 01:12:16 UTC Status MODIFIED ON_QA
Fixed In Version 4.8.0-450.ci
Sidhant Agrawal 2021-07-12 12:09:44 UTC Status ON_QA ASSIGNED
Sébastien Han 2021-07-12 12:37:45 UTC Status ASSIGNED POST
Link ID Github rook/rook/pull/8297
Sébastien Han 2021-07-12 15:00:37 UTC Link ID Github openshift/rook/pull/276
OpenShift BugZilla Robot 2021-07-12 16:13:56 UTC Status POST MODIFIED
Mudit Agarwal 2021-07-13 02:05:45 UTC Status MODIFIED ON_QA
Fixed In Version 4.8.0-450.ci 4.8.0-452.ci
Jiffin 2021-07-13 05:36:41 UTC CC jthottan
Sidhant Agrawal 2021-07-13 16:00:20 UTC Status ON_QA VERIFIED
Elad 2021-08-25 09:25:10 UTC Keywords AutomationBackLog
Red Hat Bugzilla 2022-01-10 10:25:14 UTC CC ratamir
Ramakrishnan Periyasamy 2022-08-17 10:01:39 UTC CC rperiyas
Red Hat Bugzilla 2022-12-31 19:21:10 UTC QA Contact sagrawal nberry
Red Hat Bugzilla 2022-12-31 19:35:02 UTC CC rperiyas
Red Hat Bugzilla 2022-12-31 19:54:34 UTC CC nberry
QA Contact nberry
Red Hat Bugzilla 2022-12-31 20:00:17 UTC CC jthottan
Red Hat Bugzilla 2022-12-31 22:33:42 UTC CC owasserm
Alasdair Kergon 2023-01-04 04:47:42 UTC QA Contact sagrawal
Alasdair Kergon 2023-01-04 05:01:03 UTC CC jthottan
Alasdair Kergon 2023-01-04 05:18:56 UTC CC nberry
Alasdair Kergon 2023-01-04 05:26:53 UTC CC owasserm
Alasdair Kergon 2023-01-04 05:37:14 UTC CC rperiyas
Red Hat Bugzilla 2023-01-31 23:38:20 UTC CC madam
Red Hat Bugzilla 2023-08-03 08:28:40 UTC CC ocs-bugs

Back to bug 1978722