Description of problem: When group created on a namespace of a pool, is enabled for mirroring, the renaming of the group leaves the mirrored group status in up+error state (Description: bootstrap failed) Version-Release number of selected component (if applicable): 19.2.1-138.el9cp How reproducible: Steps to Reproduce: - Create pool on primary and secondary cluster ceph osd pool create p2 - Create namespace in pool rbd namespace create p2/ns2 - Create image on primary side rbd create --size 1G p2/ns2/i1 - Enable pool for mirroring on both sides rbd mirror pool enable p2 image - Create group on namespace ns2 and add image to group rbd group create --pool p2 --group g2 --namespace ns2 rbd group image add --image i1 --group g2 --group-pool p2 --image-pool p2 --group-namespace ns2 --image-namespace ns2 - Enable namespace level mirroring on both sides rbd mirror pool enable p2/ns2 image --remote-namespace ns2 rbd mirror pool enable p2/ns2 image --remote-namespace ns2 - Create bootstrap and import on secondary rbd mirror pool peer bootstrap create --pool p2 --site-name ceph-rbd1 eyJmc2lkIjoiYjYyZWM4ZGEtMTkyNi0xMWYwLTkxOTYtZmExNjNlYWUxYzViIiwiY2xpZW50X2lkIjoicmJkLW1pcnJvci1wZWVyIiwia2V5IjoiQVFBRC8veG5hSlhDTXhBQTJaQmdCeUNlRTRCdmtrRm9Kb1lSK2c9PSIsIm1vbl9ob3N0IjoiW3YyOjEwLjAuNjQuNTc6MzMwMC8wLHYxOjEwLjAuNjQuNTc6Njc4OS8wXSxbdjI6MTAuMC42Ni40OjMzMDAvMCx2MToxMC4wLjY2LjQ6Njc4OS8wXSxbdjI6MTAuMC42Ni4yMDg6MzMwMC8wLHYxOjEwLjAuNjYuMjA4OjY3ODkvMF0ifQ== rbd mirror pool peer bootstrap import boot --pool p2 - ENable group mirroring on primary rbd mirror group enable --group g2 --pool p2 --namespace ns2 [root@ceph-rbd1-cd-cg126-dz8z92-node2 ~]# rbd mirror group status p2/ns2/g2 g2: global_id: 98a5f833-74b5-4aca-b5f2-d134c95e719b state: up+stopped description: last_update: 2025-04-18 14:38:46 images: image: 62/41d16157-b6f2-4f01-b8c7-81985aabec3e state: up+stopped description: local image is primary peer_sites: name: ceph-rbd2 state: up+replaying description: replaying last_update: 2025-04-18 14:38:34 images: image: 62/41d16157-b6f2-4f01-b8c7-81985aabec3e state: up+replaying description: replaying, {"bytes_per_second":0.0,"bytes_per_snapshot":0.0,"last_snapshot_bytes":0,"last_snapshot_sync_seconds":0,"remote_snapshot_timestamp":1744987108,"replay_state":"idle"} snapshots: .mirror.primary.98a5f833-74b5-4aca-b5f2-d134c95e719b.1e78c72a7b3bb - Rename group. The status gradually transitions into up+error state [root@ceph-rbd1-cd-cg126-dz8z92-node2 ~]# rbd group rename --pool p2 --group g2 --dest-pool p2 --dest-group g2_new --namespace ns2 --dest-namespace ns2 root@ceph-rbd2-cd-cg126-dz8z92-node2 ~]# rbd mirror group status p2/ns2/g2_new g2_new: global_id: 98a5f833-74b5-4aca-b5f2-d134c95e719b state: up+error description: bootstrap failed service: ceph-rbd2-cd-cg126-dz8z92-node5.wnxcfv on ceph-rbd2-cd-cg126-dz8z92-node5 last_update: 2025-04-18 14:42:31 images: peer_sites: name: ceph-rbd1 state: up+stopped description: last_update: 2025-04-18 14:42:14 images: image: 58/41d16157-b6f2-4f01-b8c7-81985aabec3e state: up+stopped description: local image is primary [root@ceph-rbd2-cd-cg126-dz8z92-node2 ~]# rbd mirror group status p2/ns2/g2 rbd: failed to get mirror info for group: (2) No such file or directory Actual results: The status of the renamed mirror group shows up+error state with description as "bootstrap failed" Expected results: The status of the renamed mirror group should show same as that of the status before renaming. Additional info:
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Important: Red Hat Ceph Storage 8.1 security, bug fix, and enhancement updates), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2025:9775