[DRBD-user] DRBD9 and Proxmox - no quorum in 2 nodes cluster

Shafeek Sumser shafeeks at gmail.com
Mon Mar 13 10:37:39 CET 2017

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.


Hello drbd community,

I am currently installing drbd9 with Proxmox 4.4.

The architecture is as follows:
- 2 servers (pve1 & pve2) cluster for Proxmox and a PC (pve3) to establish
the quorum for proxmox cluster.
- On the 2 servers (pve1 & pve2), drdb9 is installed

On pve1, I initialised the drbd9 cluster and I added the pve2 (drbdmanage
add-node command) to form the DRBD9 cluster.
I created a virtual machine 100 on pve1 and the volume vm-100-disk1 was
created.

The results drbd-overview on nodes pve1 & pve2 are as below:

root at pve1:~# drbd-overview
  0:.drbdctrl/0      Connected(2*) Primar/Second UpToDa/UpToDa
  1:.drbdctrl/1      Connected(2*) Primar/Second UpToDa/UpToDa
100:vm-100-disk-1/0  Connected(2*) Primar/Second UpToDa/UpToDa

root at pve2:~# drbd-overview
  0:.drbdctrl/0      Connected(2*) Second/Primar UpToDa/UpToDa
  1:.drbdctrl/1      Connected(2*) Second/Primar UpToDa/UpToDa
100:vm-100-disk-1/0  Connected(2*) Second/Primar UpToDa/UpToDa

The following scenario was tested:

1. When all nodes are running, I can migrate the virtual machine (both
online and offline) from pve1 to pve2 and from pve2 to pve1.

2. When the virtual machine is on pve2 and I shutdown the server pve2, I am
able to move the container manually from pve2 the node pve1 and start it.
This works well and the volume vm-100-disk1 changed its state from the
secondary state to primary.

3. When the virtual machine is on pve1 and I shutdown the server pve1, I am
able to move the container from node pve1 to pve2 but unable to start it.
Moreover, the drdb storage becomes unavailable and the drbd quorum is lost.

The log below error was found in the log:
Mar 11 21:52:19 pve2 pvedaemon[1291]: <root at pam> starting task
UPID:pve2:00001695:00022F2F:58C43953:vzstart:100:root at pam:
Mar 11 21:52:19 pve2 pvedaemon[5781]: starting CT 100:
UPID:pve2:00001695:00022F2F:58C43953:vzstart:100:root at pam:
Mar 11 21:52:19 pve2 drbdmanaged[1340]: ERROR      Unable to append ([122,
'Startup not successful (no quorum? not *both* nodes up in a 2 node
cluster?)', []],) to message with signature a(isa(ss))a(ssa{ss}a(ia{ss})):
<type 'exceptions.TypeError'>: 'int' object is not iterable
Mar 11 21:52:24 pve2 pvedaemon[5781]: command 'lxc-start -n 100' failed:
exit code 1

If don't know if there is a mistake in my configuration, but it works in
one way only.

If you came across such problem, your help, guide and solutions will be
much appreciated.

Thanks

Best Regards

-- 
Shafeek SUMSER
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20170313/bd96fd2f/attachment.htm>


More information about the drbd-user mailing list