[Pacemaker] Problem with SBD

vkrav at freenet.de vkrav at freenet.de
Sun Dec 12 14:37:03 UTC 2010


Hello,
I'm sorry, but I didn't found any mail list for SBD only, therefore I try to post it here:
we’ve got unfortunately big problems with SBD. We use two nodes cluster (HP-Blades) mit SBD-device as quorum on the third server (open-iscsi-server, VMWare-Server). These two nodes connect to the open-iscsi-server via open-iscsi. We have to set the watchdog timeout to 420 sec. In case we switch the open-iscsi-server off, we found out that after 120-130 sec. both nodes reset (why not after 420 sec?). We checked SBD without corosync and pacemaker – the same behavior. We checked open-iscsi without running SBD –both nodesrun without reset. As you can see the problem is only SBD.
Are there any limits in SBD? In the SBD-Wiki there is no word about that.
Just for info:
bash#:/root>sbd -d /dev/sda dump
Header version : 2
Number of slots : 255
Sector size : 512
Timeout (watchdog) : 420
Timeout (allocate) : 2
Timeout (loop) : 1
Timeout (msgwait) : 10

bash#:/root>crm configure show | grep -A 1 sbd
primitive st_sbd stonith:external/sbd \
 params sbd_device="/dev/sda" \
 op monitor interval="15s" timeout="840s"
--
clone sbd_clone st_sbd \
 meta target-role="Started"

Thank you in advance
Best regards 
Vlad




---
Exklusiv: Neue E-Mail-Adresse @iPhone.de jetzt verfügbar!
Sichern Sie sich jetzt ihre persönliche http://www.iphone.de/iphonemail/index.html?pid=10111947021



More information about the Pacemaker mailing list