Hello,<br><br>I'm configuring an HA Hadoop cluster using DRBD and Heartbeat. Works OK except failover causes reboot due to:<br><div style="margin-left: 40px;"><br> /etc/ha.d/resource.d/drbddisk r0 stop returning error: State change failed: (-12) Device is held open by someone.<br>
</div><br>from ha.debug log.<br><br>ha.conf:<br>## start of <a href="http://ha.cf">ha.cf</a><br>debugfile /var/log/ha.debug<br>logfile /var/log/ha.log<br>logfacility local0<br>keepalive 1<br>initdead 60<br>deadtime 5<br>mcast eth0 239.0.0.22 694 1 0<br>
node master1.mydomain.prod master2.mydomain.prod<br>auto_failback off<br>## end of <a href="http://ha.cf">ha.cf</a><br><br>haresources:<br>## start of haresources<br>master1.inadco.prod IPaddr::10.17.133.57<br>master1.inadco.prod drbddisk::r0<br>
master1.inadco.prod Filesystem::/dev/drbd0::/hadoop::ext3::defaults<br>master1.inadco.prod hadoop-0.20-namenode<br>master1.inadco.prod hadoop-0.20-secondarynamenode<br>master1.inadco.prod hadoop-0.20-jobtracker<br>## end of haresources<br>
<br>Any ideas?<br><br>thx,<br clear="all"><br>-- <br>Sean Barden<br>Inadco<br><a href="mailto:sbarden@inadco.com">sbarden@inadco.com</a><br><a href="http://www.inadco.com">www.inadco.com</a><br><br>