Potentially the same effect with Kernel 2.6.32-bpo.5-xen-amd64 DRBD on top of LVM on top of a RAID1-Volume 2 SATA-Disks Fujitsu Hardware one machine with 4, the other with 8 cores running XEN
This makes me fear that upgrading to Squeeze would not solve the problem. Removing one of the mirrors, thus radically avoiding check/resync worked around (or at least soothed) the problem. Interestingly enough, the hung has never occured on the respective DRBD peers, only on the machines where the heavy loaded VM is actually running on. Best regards, Tom -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4d9ecf09.2030...@graeff.com