[PVE-User] ceph could not work with PVE2.1 problem

[Alexandre DERUMIER]

>>From the PVE GUI, when I create one VM, and setting the "Hard disks"
>>of  one new VM, I saw the  "Avail" and "Capactivy" of rdb pool are all
>>zero. Is it not correct for my ceph cluster settings?
>>I just want to check the ceph cluster whether is ready for test
>>environment. The method I thought is to use the host of PVE cmd line
>>to check that. Is there any other method? thanks

This is because they are no method to retrieve the free space from rbd cluster with rbd command.

I'll try find a way in the coming week.
(one way is to use parse "ceph -w status", but don't know if we can do it remotely, and also if we can have the free space by pool )


----- Mail original ----- 

De: "para zheng" <para.pve at gmail.com> 
À: "Alexandre DERUMIER" <aderumier at odiso.com> 
Cc: pve-user at pve.proxmox.com 
Envoyé: Lundi 15 Octobre 2012 09:03:51 
Objet: Re: [PVE-User] ceph could not work with PVE2.1 problem 

2012/10/15 Alexandre DERUMIER <aderumier at odiso.com>: 
> Hi, the ceph/rbd cluster is only available from kvm userspace. 
> 
> So you can't see it from your host. 
> 
> 
> 
> if you wan't to use rbd command (why?), you need pass arguments to your command line. 

>From the PVE GUI, when I create one VM, and setting the "Hard disks" 
of one new VM, I saw the "Avail" and "Capactivy" of rdb pool are all 
zero. Is it not correct for my ceph cluster settings? 
I just want to check the ceph cluster whether is ready for test 
environment. The method I thought is to use the host of PVE cmd line 
to check that. Is there any other method? thanks 

> here the proxmox code. 
> my $cmd = ['/usr/bin/rbd', '-p', $scfg->{pool}, '-m', $monhost, '-n', 
> "client.$scfg->{username}", 
> '--keyring', "/etc/pve/priv/ceph/${storeid}.keyring", 
> '--auth_supported', $scfg->{authsupported}, $op]; 
> 
> ----- Mail original ----- 
> 
> De: "para zheng" <para.pve at gmail.com> 
> À: "Alexandre DERUMIER" <aderumier at odiso.com> 
> Cc: pve-user at pve.proxmox.com 
> Envoyé: Lundi 15 Octobre 2012 07:59:42 
> Objet: Re: [PVE-User] ceph could not work with PVE2.1 problem 
> 
> After changing version to pve-test, I can see the "ceph" in the web, 
> but the capacity is empty... 
> If i use command line to login to the PVE server, how can I check the 
> ceph cluster storage in the PVE server by command? such as "df" or 
> others command? 
> 
> I use "df", but it seems no result. 
> # df -h 
> Filesystem Size Used Avail Use% Mounted on 
> /dev/mapper/pve-root 7.9G 2.0G 5.5G 27% / 
> tmpfs 1000M 0 1000M 0% /lib/init/rw 
> udev 990M 200K 990M 1% /dev 
> tmpfs 1000M 44M 956M 5% /dev/shm 
> /dev/mapper/pve-data 16G 827M 15G 6% /var/lib/vz 
> /dev/sda1 495M 78M 392M 17% /boot 
> /dev/fuse 30M 20K 30M 1% /etc/pve 
> 
> when i use "rbd ls", there is still the same error. 
> # rbd ls 
> 2012-10-15 13:28:10.194565 7f28f427b760 -1 did not load config file, 
> using default settings. 
> unable to find any monitors in conf. please specify monitors via -m 
> monaddr or -c ceph.conf 
> error: couldn't connect to the cluster! 
> 
> I had cp the keyring file to the following path. 
> # ls /etc/pve/priv/ceph/mycephcluster.keyring 
> /etc/pve/priv/ceph/mycephcluster.keyring 
> 
> cat the conf file: 
> # cat /etc/pve/storage.cfg 
> dir: local 
> path /var/lib/vz 
> shared 
> content images,iso,vztmpl,rootdir 
> maxfiles 1 
> 
> rbd: mycephcluster 
> monhost 192.168.75.194:6789;192.168.75.195:6789;192.168.75.196:6789 
> pool rbd 
> authsupported cephx;none 
> username admin 
> content images 
> 
> thanks