TOPIC:

So, I'm doing a PoC test for HA-Lizard/iSCSI-HA in my lab. Here's the topology:

Switch: Cisco 3750G, 24 Port 1Gbe switch
Hosts: 2xWhitebox servers; supermicro Motherboards, 2xIntel Xeon CPU, 64GB RAM, 500GB SATA boot drive, Dell MD1000e enclosure directly connected to an LSI controller.

Network: 1xManagement, 1xReplication for testing. Quorum IP is the IP of the management SVI.

After initial installation, everything worked wonderfully. I set up the pool per the documentation, iscsi-ha is happy, replication completed successfully. I created an SR using the virtual IP assigned to the DRBD cluster, and created a Server 2008 R2 VM.

Since that was successful, I induced a failure by cutting the power to the pool master. My VM was running on the other host. 2nd host took over, VM never shut down. Everything is golden. I powered host 1 back on, and I have a reboot loop. The server will not completely start up. It gets to a point of loading the autofs module, displays the XenServer Splash screen, the progress bar move across, then the host reboots.
Any idea why this is happening?

I'm pretty sure this was fixed in the latest version 2.1.4. What version of halizard are you running?

iscsi-ha version 2.1.4, ha-lizard version 2.1.3. I didn't see 2.1.4 on the website?

You can download directly from link below. I'll have our website updated in the meantime.
Thanks

halizard.org/release/ha-lizard/

Thanks. Any way I can boot this system to upgrade it, or am I going to have to wipe it and start from scratch?

If you can manage to get into the console for a few seconds before a reboot occurs, paste the following 3 lines into the console. It may reboot once more and then stop

> /etc/ha-lizard/state/fenced_slave && sync
chkconfig ha-lizard-watchdog off
chkconfig ha-lizard off

After upgrading
chkconfig ha-lizard-watchdog on
chkconfig ha-lizard on
service ha-lizard-watchdog start