FYI: My system died on FS corruption. Did not notice this because mdadm did not report any errors. I noticed it because the system started to lock up. After a reboot, it did a fsck (which segfaulted) and the system hung afterwards.
The system behaviour seems like this kernel bug describes: http://bugzilla.kernel.org/show_bug.cgi?id=11328 mdadm reports both arrays as healthy, but secretly all data was being corrupted. Maybe add a reference to this bug? I know this bug is directed at the HPT driver, however due to the similarities in the problems it could be a bigger, underlying, problem. I'm going to try and rebuild the system using a 2.6.24 kernel since that is the last kernel which did not report data corruption problems when using with mdadm. Rudi -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
