Le 2025-08-21 16:11, Gilles Mocellin a écrit :
Ok,
It happens on only one host, and ti was the DB device that certainly
holds some information.
I don't konw what, it had room for new LVs...
I zap it, with the 4 OSDs that where using it.
And new OSDs are now created with their DB device.
Argh !
I was wrong, it only creates correctly the OSD that were previously with
that DB device.
Th other OSD that didn't have a DB device are recreated without DB
device.
I delete them with :
ceph orch osd rm ID --zap
It's like if it works only the first time we apply a spec, with an empty
DB device.
When the DB device has some LV, it's not anymore took into account...
I've put the OSD spec in unmanaged mode, drain my host, stoping and
removing locally (cephadm ceph-volume lvm zap...) 3 OSDs that were not
drained.
Then, an apply --dry-run of my OSD spec shows that it will use my
db-device.
I applied it, an d yes, it's used for every OSD on that host.
Now, what is not a normal behavior, is that if we remove/replace an OSD,
it should be recreated with a DB device.
It's actually not (squid 19.2.3).
Tested on several hosts.
_______________________________________________
ceph-users mailing list -- ceph-users@ceph.io
To unsubscribe send an email to ceph-users-le...@ceph.io
