Dear Ceph users,I noticed that updating a Rocky Linux 9 host with dnf in my cluster results in some of the Cephadm daemons vanishing. Before dnf update:
$ sudo systemctl list-units | grep cephceph-b1029256-7bb3-11ec-a8ce-ac1f6b627b45@xxxxxxxxxxxxxxxxx.service loaded active running Ceph alertmanager.kili for b1029256-7bb3-11ec-a8ce-ac1f6b627b45 ceph-b1029256-7bb3-11ec-a8ce-ac1f6b627b45@xxxxxxxxxx.service loaded active running Ceph crash.kili for b1029256-7bb3-11ec-a8ce-ac1f6b627b45 ceph-b1029256-7bb3-11ec-a8ce-ac1f6b627b45@xxxxxxxxxxxx.service loaded active running Ceph grafana.kili for b1029256-7bb3-11ec-a8ce-ac1f6b627b45 ceph-b1029256-7bb3-11ec-a8ce-ac1f6b627b45@xxxxxxxxxxxxxxxxxx.service loaded active running Ceph node-exporter.kili for b1029256-7bb3-11ec-a8ce-ac1f6b627b45 ceph-b1029256-7bb3-11ec-a8ce-ac1f6b627b45@osd.117.service loaded active running Ceph osd.117 for b1029256-7bb3-11ec-a8ce-ac1f6b627b45 ceph-b1029256-7bb3-11ec-a8ce-ac1f6b627b45@osd.118.service loaded active running Ceph osd.118 for b1029256-7bb3-11ec-a8ce-ac1f6b627b45 ceph-b1029256-7bb3-11ec-a8ce-ac1f6b627b45@osd.119.service loaded active running Ceph osd.119 for b1029256-7bb3-11ec-a8ce-ac1f6b627b45 ceph-b1029256-7bb3-11ec-a8ce-ac1f6b627b45@osd.120.service loaded active running Ceph osd.120 for b1029256-7bb3-11ec-a8ce-ac1f6b627b45 ceph-b1029256-7bb3-11ec-a8ce-ac1f6b627b45@osd.121.service loaded active running Ceph osd.121 for b1029256-7bb3-11ec-a8ce-ac1f6b627b45 ceph-b1029256-7bb3-11ec-a8ce-ac1f6b627b45@osd.122.service loaded active running Ceph osd.122 for b1029256-7bb3-11ec-a8ce-ac1f6b627b45 ceph-b1029256-7bb3-11ec-a8ce-ac1f6b627b45@osd.123.service loaded active running Ceph osd.123 for b1029256-7bb3-11ec-a8ce-ac1f6b627b45 ceph-b1029256-7bb3-11ec-a8ce-ac1f6b627b45@osd.62.service loaded active running Ceph osd.62 for b1029256-7bb3-11ec-a8ce-ac1f6b627b45 system-ceph\x2db1029256\x2d7bb3\x2d11ec\x2da8ce\x2dac1f6b627b45.slice loaded active active Slice /system/ceph-b1029256-7bb3-11ec-a8ce-ac1f6b627b45 ceph-b1029256-7bb3-11ec-a8ce-ac1f6b627b45.target loaded active active Ceph cluster b1029256-7bb3-11ec-a8ce-ac1f6b627b45 ceph.target loaded active active All Ceph clusters and services
After dnf update: $ sudo systemctl list-units | grep cephceph-b1029256-7bb3-11ec-a8ce-ac1f6b627b45@xxxxxxxxxxxxxxxxxx.service loaded active running Ceph node-exporter.kili for b1029256-7bb3-11ec-a8ce-ac1f6b627b45 system-ceph\x2db1029256\x2d7bb3\x2d11ec\x2da8ce\x2dac1f6b627b45.slice loaded active active Slice /system/ceph-b1029256-7bb3-11ec-a8ce-ac1f6b627b45 ceph-b1029256-7bb3-11ec-a8ce-ac1f6b627b45.target loaded active active Ceph cluster b1029256-7bb3-11ec-a8ce-ac1f6b627b45
ceph.targetI guess this is due to dnf updating several Docker packages (docker-ce, docker-ce-cli etc.), but the same does not occur on the Rocky Linux 8 hosts, where after system updates all the Cephadm daemons are still running.
Does anybody have an idea about why this might happen? Thanks, Nicola
Attachment:
smime.p7s
Description: S/MIME Cryptographic Signature
_______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
