Skip to content

mover: fetch ceph-csi-config via k8s API instead of volume mount #23

Merged
Rakshith-R merged 2 commits into
RamenDR:mainfrom
Rakshith-R:csv-changes
Apr 15, 2026
Merged

mover: fetch ceph-csi-config via k8s API instead of volume mount #23
Rakshith-R merged 2 commits into
RamenDR:mainfrom
Rakshith-R:csv-changes

Conversation

@Rakshith-R

@Rakshith-R Rakshith-R commented Apr 15, 2026

Copy link
Copy Markdown
Collaborator

The operator no longer mounts the ceph-csi-config ConfigMap as a volume.
Instead, it fetches it via the Kubernetes API in the reconcile loop using
a shared fetchCSIConfigData helper. The ConfigMap name and namespace are
configurable via CEPH_CSI_CONFIG_NAME and CEPH_CSI_CONFIG_NAMESPACE env
vars (defaults: ceph-csi-config / rook-ceph), with Makefile placeholder
substitution matching the existing MOVER_IMAGE pattern.

Mover Job pods remain unchanged - they still mount their per-RS/RD copy.

Also increases operator resource limits (cpu: 1000m, memory: 512Mi) and
adds VolSync CRD dependencies to the ClusterServiceVersion.

Assisted-by: Claude noreply@anthropic.com
Signed-off-by: Rakshith R rar@redhat.com

@Rakshith-R Rakshith-R marked this pull request as draft April 15, 2026 06:10
The operator no longer mounts the ceph-csi-config ConfigMap as a volume.
Instead, it fetches it via the Kubernetes API in the reconcile loop using
a shared fetchCSIConfigData helper. The ConfigMap name and namespace are
configurable via CEPH_CSI_CONFIG_NAME and CEPH_CSI_CONFIG_NAMESPACE env
vars (defaults: ceph-csi-config / rook-ceph), with Makefile placeholder
substitution matching the existing MOVER_IMAGE pattern.

Mover Job pods remain unchanged - they still mount their per-RS/RD copy.

Also increases operator resource limits (cpu: 1000m, memory: 512Mi) and
adds VolSync CRD dependencies to the ClusterServiceVersion.

Assisted-by: Claude <noreply@anthropic.com>
Signed-off-by: Rakshith R <rar@redhat.com>
setRSPaused, updateManualTrigger, and updateRSManualTrigger do a
Get+Update without retry. When the controller updates the resource
between those calls, the test hits a 409 Conflict and fails. Wrap
each Get+Update in retry.RetryOnConflict to handle the race.

Assisted-by: Claude <noreply@anthropic.com>
Signed-off-by: Rakshith R <rar@redhat.com>
@Rakshith-R Rakshith-R changed the title build: enable OwnNamespace install mode and add volsync CRD dependencies mover: fetch ceph-csi-config via k8s API instead of volume mount Apr 15, 2026
@Rakshith-R Rakshith-R marked this pull request as ready for review April 15, 2026 12:39
@Rakshith-R Rakshith-R merged commit 2e6efe1 into RamenDR:main Apr 15, 2026
6 of 7 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant