<html><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><br><div><div>On Nov 19, 2009, at 3:03 PM, Andrew Beekhof wrote:</div><blockquote type="cite"><div><font class="Apple-style-span" color="#000000"><br></font><blockquote type="cite">Another problem has appeared:<br></blockquote><blockquote type="cite">after the reboot of one server I often have a cluster partition and both<br></blockquote><blockquote type="cite">servers elect themselves DC.<br></blockquote><blockquote type="cite">Even if the partition doesn't appear just after the reboot of one server<br></blockquote><blockquote type="cite">(i.e. serverA), if I try to restart corosync on the other server (i.e.<br></blockquote><blockquote type="cite">serverB), the partition appear.<br></blockquote><blockquote type="cite">Then if I also restart corosync on the first server (serverA) everything<br></blockquote><blockquote type="cite">work fine again.<br></blockquote><blockquote type="cite">But if I restart corosync on the second server (serverB) nothing change and<br></blockquote><blockquote type="cite">the partition appears again.<br></blockquote><blockquote type="cite">It's seems to me that there is still something wrong with the first run of<br></blockquote><blockquote type="cite">corosync just after the server reboot.<br></blockquote><br>I've found that it starts a bit too early by default.<br>Various systems seem to like messing with the network stack (xen is<br>one but there are others) which confuses corosync.<br></div></blockquote><div><br></div>I wrote a shell script that "manually starts" corosync 5 minutes after the server starts and in this case the problem appears every time!</div><div>It's driving me crazy, because I can see that my script starts a while after the server is up and I'm pretty sure everything is running!</div><div>On the other hand, if I start manually corosync just after the server is up, everything works fine!</div><div><div><br></div><br><blockquote type="cite"><div>You're not getting addresses from a dhcp server are you?<br>Thats another common cause, since there can be a significant delay in<br>obtaining the address - which again messes with corosync.<br></div></blockquote><div><br></div>Absolutely no!</div><div>I have two servers with static public IP.</div><div>I also added the two server in the /etc/hosts file: in general I followed all the guidelines I found in the documentation.</div><div><div><br></div><br><blockquote type="cite"><div><blockquote type="cite">I didn't configure any fencing method, because I think that my configuration<br></blockquote><blockquote type="cite">is really simple and I don't need it.<br></blockquote><br>Do you need your data though?<br></div></blockquote></div><div><br></div><div>Do you mean it's better to configure a fencing method anyway?</div><div><br></div><div>Thank you very much for your help!</div><div>Giovanni</div></body></html>