Conversion to CEPH Complete

M

mudmarine

New Member
Mar 12, 2016
2
0
1
50
Just wanted to drop the Proxmox folks note as well as anyone else interested, in the fact that I have completely ditched LVM, Mergerfs, RAIDx, ZFS, soft raid, etc!!!!!!!!:D

I am running a single node ceph cluster on my home server box with a pair of 1tb drives. I can mount them into my containers and VMs from the GUI and all is right in the world now. Ceph is the future. Dump all that old tech and get on board with a big iron file system.

Currently running Ceph 0.94.6. Hoping that proxmox folks get it updated to latest in their next patch. I would do it myself, but don't want to trash anything.
 
mudmarine said:
J
Currently running Ceph 0.94.6. Hoping that proxmox folks get it updated to latest in their next patch. I would do it myself, but don't want to trash anything.
Click to expand...

Hammer is ok. For jewel, the debian packaging (from ceph.com) have some missing systemd files and other things.
So it's better to wait a little bit.
 
I tried manually moving to jewel on a test system and it completely trashed things. Hold off for now.
 
mudmarine said:
I tried manually moving to jewel on a test system and it completely trashed things. Hold off for now.
Click to expand...
Hi,
what do you mean with "completly trashed things"?

I test the upgrade to jewel on an test-pve-system too and it's works with some limitations.

1. you have to add the user ceph to the group www-data (because of the right-management in /etc/pve)
Code: 
addgroup ceph www-data
2. you have to mount the osds by yourself, because the start-scripts are broken (use the right mount-option)

3. You have to change the permission for user ceph one time
Code: 
chown -R ceph:ceph /var/lib/ceph/osd
chown -R ceph:ceph /var/lib/ceph/mon/

4. you have to start ceph (mon+osds) manually because of broken start-scripts too
Code: 
sudo -u ceph /usr/bin/ceph-mon -i 0 --pid-file /var/run/ceph/mon.0.pid -c /etc/ceph/ceph.conf --cluster ceph
sudo -u ceph /usr/bin/ceph-osd -i 2 --pid-file /var/run/ceph/osd.2.pid -c /etc/ceph/ceph.conf --cluster ceph

Work for me, but this is for testing only! Not for production, of course.

Udo
 
You must log in or register to reply here.
Top Bottom
Back