Hi! I wonder: Are you trying to create a snapshot of a VG that is activated on another node? A snapsot modifies the data blocks of the original VG, so I'd be not surprised if the kernel crashes then.
Regards, Ulrich >>> Brent Clark 06.05.11 11.44 Uhr >>> Hiya Im wondering if someone could share some thought on a problem that my colleague and I are experiencing. Where I work, we got this really weird problem whereby any servers in a cluster pair may on occasion reboot. Im thinking its due to high IO. But I cant prove it. We have systat installed and via sar, nothing really sticks out as to what the culprit may be. The setup of and between the two servers is : Heartbeat, Drbd (Active - Active) for the websites content and then MySQL replication (Active - Passive). There is a second nic on both machines and via the backlink / cross over cable, all activity happens. Both servers make use of Software Raid 1. Its cant be hardware, as we have replaced the it (except the drives). The Os is Debian Lenny 32Bit. One of the key things is, we make use of mylvmbackup, and whats interesting is that once the server has rebooted, and we get the Nagios notification, on logging in, we see the temporary LV of mylvmbackup. And what interesting is, that both servers get affected. So the purpose of HA is kinda defeated. :) I originally thought it may be heartbeat that is the culprit, as I know heartbeat has the ability to reboot, but we cant find anything in the logs. If anyone could share a thought or may have I idea or where more to look, it would gratefully be appreciated. Kind Regards Brent Clark P.s. We only run mylvmbackup on the passive server. _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
