I have a three node setup all running Ubuntu 12.04 and DRBD 8.3.11. I have the drbd service configured to not startup automatically. Instead I launch it with post-up commands in /etc/network/interfaces via modprobe drbd followed by drbdadm up all. This works fine. All three nodes startup and resync after a reboot. But they are all in the secondary role as expected. On my primary node I have another interface defined that I use as an NFS target for my ESXi hosts. I do this so if something happened to the primary node I could manually bring up the secondary node in its place. The first post-up command I have is drbdadm primary all. Unfortunately when I reboot the server something about this causes the boot process to hang. It claims it is waiting for the network to start up. While it is doing this I can connect into the server just fine remotely but in the end it doesn't execute any of the post-up commands for eth1. However, I can run each command individually from the command line and they completely quickly and without error. So for now I have removed those post-up commands and put them in a script I manually run when I reboot the server. Of course this does me no good if the server is rebooted for some other reason.<div>
<br></div><div>Looking through the syslog I can't seem to find any indication why the drbdadm primary all would fail when it's run at boot time. Any suggestions?<br clear="all"><div><br></div>-- <br>J.R. Lillard<div>
System / Network Admin</div><div>Web Programmer</div><div>Golden Heritage Foods</div><div>120 Santa Fe St.</div><div>Hillsboro, KS 67063</div><br>
</div>