2

I currently am trying to make a highly available NAS under debian 7. I am using following packages:

drbd8-utils
iscsitarget
iscsitarget-dkms
pacemaker
corosync
openais

everything works fine without pacemaker (so I'll take it's not a conf error)

When I put everything on pacemaker, everything works until I reboot one of the nodes. here is my pacemaker config :

node NAS-1
node NAS-2
primitive nas_ip ocf:heartbeat:IPaddr2 \
    params ip="192.168.152.210" cidr_netmask="255.255.255.0"
primitive p_drbd ocf:linbit:drbd \
    params drbd_resource="nas" \
    op monitor interval="29" role="Master" \
    op monitor interval="31" role="Slave"
primitive p_iscsitarget ocf:heartbeat:iSCSITarget \
    params iqn="iqn.2015.01.net.utopia:nas" tid="1" \
    op monitor interval="10s"
group g_iscsi nas_ip p_iscsitarget
ms ms_drbd p_drbd \
    meta master-max="1" master-node-max="1" clone-max="2" clone-node-max="1" notify="true"
colocation nas inf: g_iscsi ms_drbd:Master
order o_drbd inf: ms_drbd:promote g_iscsi:start
property $id="cib-bootstrap-options" \
    dc-version="1.1.7-ee0730e13d124c3d58f00016c3376a1de5323cff" \
    cluster-infrastructure="openais" \
    expected-quorum-votes="2" \
    stonith-enabled="false" \
    no-quorum-policy="ignore" \
    default-resource-stickiness="1"
crm(live)configure#

Here is the following error :

root@NAS-1:~# crm_mon -1
============
Last updated: Wed Jan  7 16:02:53 2015
Last change: Wed Jan  7 15:57:51 2015 via cibadmin on NAS-1
Stack: openais
Current DC: NAS-2 - partition with quorum
Version: 1.1.7-ee0730e13d124c3d58f00016c3376a1de5323cff
2 Nodes configured, 2 expected votes
4 Resources configured.
============
Online: [ NAS-1 NAS-2 ]

Master/Slave Set: ms_drbd [p_drbd]
    Masters: [ NAS-2 ]
    Slaves: [ NAS-1 ]
Resource Group: g_iscsi
    nas_ip     (ocf::heartbeat:IPaddr2):       Started NAS-2
    p_iscsitarget      (ocf::heartbeat:iSCSITarget):   Started NAS-2 (unmanaged) FAILED

Failed actions:
    p_iscsitarget_stop_0 (node=NAS-2, call=16, rc=5, status=complete): not installed
root@NAS-1:~#

Note: there is indeed not the iSCSi Logical Unit configured in the NAS, currently in testing phase on VMs, but I had the exact same error with it on my previous test.

Could someone help me pinpoint/correct that error?

PS : My conf files for other things :
http://pastebin.com/rUZYrhxm (/etc/drbd.d/*)
http://pastebin.com/X4VGNJuP (/etc/iet/ietd.conf)

3 Answers 3

1

Problem doesn't come from drbd fencing, when i reboot one node, both nodes indeed switch their roles (rebooted server becomes secondary if it wasn't already, and the other server becomes primary if it wasn't already).

I am already using resource level fencing with Cluster Information Base (see http://pastebin.com/rUZYrhxm ).

Strategy was grouping the iSCSI service startup with the drbd Master role. except that it seemed it was trying to start up iSCSITarget before switching to Master role, thus was locked since disk was in readonly.

I've redone my whole system from scratch once more, and now it works, guess i made one small error somewhere.

0

Because you are missing stonith, you need to configure stonith in pacemaker and configure drbd to call pacemaker fencing.

http://drbd.linbit.com/users-guide/s-pacemaker-fencing.html
0
0

I think is because in Debian, don't have iSCSITarget for heartbeat. Check the directory /usr/lib/ocf/resource.d/redhat/.

ls -ls /usr/lib/ocf/resource.d/redhat/

If not see there so you are calling an script that not exists.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .