________________________________ From: [email protected] on behalf of Igor Chudov Sent: Mon 7/26/2010 7:16 AM To: [email protected] Subject: [Linux-HA] Big progress with Heartbeat,but simultaneous reboot leaves services unprovided
I am setting up a two node cluster, using drbd and Heartbeat. I use standard packages on Ubuntu Hardy. The services being provided externally is a NFS and samba share that is on top of the DRBD filesystem, and the service IP address. I am not using corosync at the moment. At this point, most things work great: the shared services and IP address are passed around when servers reboot or are unplugged, etc. However, I HAVE ONE PROBLEM: if I simultaneously reboot both servers by typing reboot in both sessions, and then hitting ENTER in both at about the same time, neither of the servers acquires shared services, so they remain unprovided. If, after that, I reboot one of the servers again, then the unrebooted one acquires services. What exactly am I doing wrong? Here is my ha.cf and haresources: ==>cat ha.cf use_logd on udpport 12694 keepalive 1 warntime 15 deadtime 20 debug 1 initdead 60 bcast eth1 node pfs-srv3 node pfs-srv4 auto_failback on crm off ==>cat haresources pfs-srv3 drbddisk::r0 Filesystem::/dev/drbd0::/pfs::ext3 10.1.8.45/24 nfs-kernel-server smbd Did you see the logs? Does HA try to start the resources on the preferred node? Can you check what is the status reported by HB script (/etc/init.d/heartbeat status)? Also can you run cl_status with various arguments e.g. nodestatus, hbstatus? You can also run the individual resource scripts with the status argument to check what it reports (started/stopped)? pushkar
<<winmail.dat>>
_______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
