Q: 2 node cluster - no HA - quorum iSCSI target - issues / help request

[fortechitsolutions]

Hi, I wonder if anyone can comment.

I am trying to setup a modest size proxmox cluster,

-- 2 physical nodes running proxmox
-- to avoid problems with Quorum if one node is offline, I wanted to use a small dedicated shared iSCSI target as the 'quorum disk' for the 3rd vote 'tie breaker'.

-- to achieve this goal, basic steps I try:

-- setup cluster with 2 members, normal method as per https://pve.proxmox.com/wiki/Proxmox_VE_2.0_Cluster
-- then setup iSCSI target quorum disk and proceed with setup "HA" config starting from this step:
https://pve.proxmox.com/wiki/Two-Node_High_Availability_Cluster#Quorum_Disk_Configuration

I am NOT interested in HA and fencing. I just want ability for quorum to work when just one physical host is online, plus the Quorum disk (ie, and second/other host is offline).

But I am getting feeling the config is not supported / or I am doing something wrong.

After I carefully follow the steps, I seem to get output which is expected, ie;

root@proxmox-a:/var/log# pvecm s
Version: 6.2.0
Config Version: 3
Cluster Name: newcluster
Cluster Id: 45236
Cluster Member: Yes
Cluster Generation: 96
Membership state: Cluster-Member
Nodes: 2
Expected votes: 3
Quorum device votes: 1
Total votes: 3
Node votes: 1
Quorum: 2
Active subsystems: 3
Flags:
Ports Bound: 0 178
Node name: proxmox-a
Node ID: 1
Multicast addresses: 239.192.176.101
Node addresses: 10.82.141.48
 
root@proxmox-a:/var/log# cat /etc/pve/cluster.conf
<?xml version="1.0"?>
<cluster name="newcluster" config_version="3">

  <cman expected_votes="3" keyfile="/var/lib/pve-cluster/corosync.authkey"/>
  <quorumd votes="1" allow_kill="0" interval="1" label="proxmox1_qdisk" tko="10"/>
  <totem token="54000"/>

  <clusternodes>
  <clusternode name="proxmox-a" votes="1" nodeid="1"/>
  <clusternode name="proxmox-b" votes="1" nodeid="2"/></clusternodes>

</cluster> 

and

root@proxmox-a:/var/log# clustat
Cluster Status for newcluster @ Wed Jun 17 15:55:03 2015
Member Status: Quorate

 Member Name                                           ID   Status
 ------ ----                                           ---- ------
 proxmox-a                                                 1 Online, Local
 proxmox-b                                                 2 Online
 /dev/block/8:49                                           0 Online, Quorum Disk

root@proxmox-a:/var/log#

Alas, in web interface now if I click on the HA tab, it gives me a timeout error, "Got log request timeout (500)"

and if I try to 'touch' a file under /etc/pve it tells me, "touch: cannot touch `a': Device or resource busy"

So I think I have borked the cluster config / it thinks it has not got quorum / but I am not clear - what is the proper way to achieve this. (If it is possible ?)

Any comments are greatly appreciated.

Thanks,


Tim

 

[fortechitsolutions]

Hi, quick footnote on this, (where this info actually applies

I rebooted both hosts, just for fun, and things appear? to be working precisely as I might have hoped. I didn't change anything.

To test, I then

-- added a new NFS shared storage, and it goes in fine to config, no error-timeout (/etc/pve filesystem area not locked any more)
-- powered off one host, so we lose 2-node-only-quorum (iSCSI quorum disk still attached to the host remaining online)
-- and things are good, I have quorum still on the one online host, life is good (/etc/pve FS remains usable, I can touch:edit a file in there)

I am not sure why things were fiddley initially. I will stress test it a bit more but it seems in good order now.

Tim