[pve-devel] Following tests on PvE 4.0 Jessie

Moula BADJI moulab1 at hotmail.com
Sun May 10 08:08:59 CEST 2015


I recreated my POC . I modified the root account in sshd.conf . Cluster
creation works :

root at pve-ceph1:/home/moula# pvecm status
Quorum information
Date:             Sun May 10 06:48:01 2015
Quorum provider:  corosync_votequorum
Nodes:            3
Node ID:          0x00000001
Ring ID:          172
Quorate:          Yes

Votequorum information
Expected votes:   3
Highest expected: 3
Total votes:      3
Quorum:           2 
Flags:            Quorate

Membership information
    Nodeid      Votes Name
0x00000001          1 (local)
0x00000002          1
0x00000003          1

I noticed that :
* The command clustat does not work
* In the dashbord : Datacenter ---> Sammary : displays nothing.
* installing ksm -control- daemon never works, it crashes

I installed glusterfs for nfs, everything works. upload every iso.
installation and migration of vms work very well.

root at pve-ceph1:/home/moula# gluster volume info
Volume Name: glusterstorage
Type: Replicate
Volume ID: 1f97cb46-f985-4038-865c-cb668572cfbf
Status: Started
Number of Bricks: 1 x 3 = 3
Transport-type: tcp
Brick1: pve-ceph1:/var/lib/glusterfs
Brick2: pve-ceph2:/var/lib/glusterfs
Brick3: pve-ceph3:/var/lib/glusterfs

* I installed ceph hammer jessie but it not possible to create OSDs ,
error messages :
# pveceph createosd /dev/sdb -journal_dev dev/sdd
Use of uninitialized value $path in pattern match (m//) at
/usr/share/perl5/PVE/CephTools.pm line 50, <DATA> line 751.
Use of uninitialized value $path in concatenation (.) or string at
/usr/share/perl5/PVE/CephTools.pm line 50, <DATA> line 751.
got unusual device path ''

*next time I will test the new PVE-HA ( any howto?) and the creation of
lxc with new updates


Moula de la Kabylie.

More information about the pve-devel mailing list