Cheers Reed,
just saw this and checked on my own. Also had one mon that ran on leveldb.

I just removed the mon, pulled the new monmap and deployed it. After that
all was fine.

Thanks for paging the ML, so I've read it :D

Boris

# assuming there is only one mon and you are connected to the host
that got the mon
systemctl stop ceph-mon.target
ceph mon rm `hostname -s`
rm -rf /var/lib/ceph/mon/ceph-*
ceph mon getmap -o /tmp/monmap
sudo -u ceph ceph-mon --cluster ceph --mkfs -i `hostname -s` --monmap
/tmp/monmap
systemctl restart ceph-mon.target
rm /tmp/monmap



Am Mo., 3. Okt. 2022 um 18:37 Uhr schrieb Reed Dier <reed.d...@focusvq.com>:

> Recently reading this thread:
> https://www.mail-archive.com/ceph-users@ceph.io/msg16705.html <
> https://www.mail-archive.com/ceph-users@ceph.io/msg16705.html>
>
> And out of curiosity I decided to take a look, and it turns out, 2/3 of my
> mons are using rocksdb, while I still somehow have a leveldb mon lingering.
>
> > for host in mon1 mon2 mon3 ; do ssh -t $host sudo cat
> /var/lib/ceph/mon/ceph-$host/kv_backend ; done
> > rocksdb
> > Connection to mon1 closed.
> > rocksdb
> > Connection to mon2 closed.
> > leveldb
> > Connection to mon3 closed.
>
>
> This is on Octopus, not using cephadm.
>
> I can’t really locate any kind of conversion process in the docs or
> mailing-list for converting the kv backend from leveldb to rocksdb, short
> of blowing the mon away and re-deploying the mon from scratch.
> Which that isn’t inherently bad, but I figured that must be some other way
> to achieve this that I am apparently not finding.
>
> Hopefully someone can point me to the proper docs, or the proper
> (un(der)documented) way is to kill and replace and move on.
>
> Thanks,
> Reed
> _______________________________________________
> ceph-users mailing list -- ceph-users@ceph.io
> To unsubscribe send an email to ceph-users-le...@ceph.io
>


-- 
Die Selbsthilfegruppe "UTF-8-Probleme" trifft sich diesmal abweichend im
groüen Saal.
_______________________________________________
ceph-users mailing list -- ceph-users@ceph.io
To unsubscribe send an email to ceph-users-le...@ceph.io

Reply via email to