Cluster HA and Openvz Live migrate (and HA) doasn't work

[yena]

Hi,
i have 3 node cluster in staging; last versione.
I Cant migrate Openvz container:


Source Node
/var/log/syslog
Oct 21 12:49:31 proxmox-2 pvedaemon[359798]: <root@pam> starting task UPIDroxmox-2:0009A688:08B2C40F:526506BB:hamigrate:208:root@pam:
Oct 21 12:49:31 proxmox-2 pvedaemon[632456]: command 'clusvcadm -M pvevm:208 -m proxmox-1' failed: exit code 250
Oct 21 12:49:31 proxmox-2 pvedaemon[359798]: <root@pam> end task UPIDroxmox-2:0009A688:08B2C40F:526506BB:hamigrate:208:root@pam: command 'clusvcadm -M pvevm:208 -m proxmox-1' failed: exit code 250

tail -f /var/log/cluster/rgmanager.log
Oct 21 12:47:34 rgmanager Migrating pvevm:208 to proxmox-1
Oct 21 12:47:35 rgmanager migrate on pvevm "208" returned 1 (generic error)
Oct 21 12:47:35 rgmanager Migration of pvevm:208 to proxmox-1 failed; return code 1

Destination Node
Oct 21 12:49:31 proxmox-1 rgmanager[2889]: BUG! Attempt to forward to myself!
Oct 21 12:49:31 proxmox-1 pmxcfs[2344]: [status] notice: received log
-----------------------------------------------------

On both
pveversion
pve-manager/3.1-14/d914b943 (running kernel: 2.6.32-24-pve)

clustat
Cluster Status for testC @ Mon Oct 21 12:53:45 2013
Member Status: Quorate

Member Name ID Status
------ ---- ---- ------
proxmox-1 1 Online, rgmanager
proxmox-2 2 Online, Local, rgmanager
proxmox-3 3 Online

Service Name Owner (Last) State
------- ---- ----- ------ -----
pvevm:208 (proxmox-2) failed


pvecm s
Version: 6.2.0
Config Version: 18
Cluster Name: testC
Cluster Id: 3423
Cluster Member: Yes
Cluster Generation: 4664
Membership state: Cluster-Member
Nodes: 3
Expected votes: 2
Total votes: 3
Node votes: 1
Quorum: 2
Active subsystems: 6
Flags:
Ports Bound: 0 177
Node name: proxmox-2
Node ID: 2
Multicast addresses: 239.192.13.108
Node addresses: 192.168.0.2
-------------------------------------

I thik someting wrong on devices:
on the Source Node where VPS works i can see the device:

locate VG_DRBD0/
/dev/VG_DRBD0/openvz1

and:
root@proxmox-2:/dev/mapper# ls -la
total 0
drwxr-xr-x 2 root root 140 Oct 14 18:48 .
drwxr-xr-x 19 root root 6480 Oct 14 18:48 ..
crw------- 1 root root 10, 61 Oct 4 15:27 control
lrwxrwxrwx 1 root root 7 Oct 4 15:27 pve-data -> ../dm-2
lrwxrwxrwx 1 root root 7 Oct 4 15:27 pve-root -> ../dm-0
lrwxrwxrwx 1 root root 7 Oct 4 15:27 pve-swap -> ../dm-1
lrwxrwxrwx 1 root root 7 Oct 14 18:48 VG_DRBD0-openvz1 -> ../dm-3



On the destination node no :
i can't see: /dev/VG_DRBD0

I haven't the dm-3 devices



bu i can see it on lvdisplay on both nodes:

--- Logical volume ---
LV Path /dev/VG_DRBD0/openvz1
LV Name openvz1
VG Name VG_DRBD0
LV UUID qCmXgZ-d0tz-VCeL-Te9H-Oa84-5kYR-GzBC5P
LV Write Access read/write
LV Creation host, time proxmox-1, 2013-10-14 19:22:09 +0200
LV Status NOT available
LV Size 10.00 GiB
Current LE 2560
Segments 1
Allocation inherit
Read ahead sectors auto
----------------------------
DRBD seem OK:
version: 8.3.13 (api:88/proto:86-96)
GIT-hash: 83ca112086600faacab2f157bc5a9324f7bd7f77 build by root@sighted, 2012-10-09 12:47:51
0: cs:Connected rorimary/Primary ds:UpToDate/UpToDate C r-----
ns:0 nr:0 dw:0 dr:4612 al:0 bm:0 lo:0 pe:0 ua:0 ap:0 ep:1 wo:b oos:0
----------------------------

Some ideas ?

Thanks!

 

[dietmar]

What kind of file system do you use? Are you aware of the fact that you cannot use ext3/ext4/xfs/.. on a shared block device!

 

[yena]

I'm using ext4 ..
so to activate Migration and HA, i have to mount my logical volume in NFS or Ocfs ? right ?
Hav i to modify something in /etc/lvm/lvm.conf ?

Thanks !!

-- Yena --

 

[dietmar]

so to activate Migration and HA, i have to mount my logical volume in NFS or Ocfs ? right ?

Yes - ext4 will not work.