Hi, I've got a couple of doubts. Since more than a year ago I setup mdadm raid1 on new instalations (thanks for your work). On the older machines, mdadm -D shows them in a dirty state. I know that this is/was as expected. If there is a power cut, the machine boots and a dirty array is resynced, ¿right? On newer installations, mdadm -D shows arrays in a clean state. I had always thought that this was ok becuase I'd read somewhere that the term 'dirty' was causing confusion and it got changed. So, state: clean, is what I should expect? Last week we had a power cut and when the machines booted all arrays where brought up. I'm running debian etch and got INITRDSTART='all' AUTOSTART="all" Soon after one of the machines booted, I got users calling me saying that documents were disappearing ar becoming corrupted. I looked at cat /proc/mdstat and saw that nothing was being resynced. I stopped the array, then removed and added one of the partitions to resync it (could I have done the same using /usr/share/mdadm/checkarray ?) Then I ran fsck.ext3 on it (the /dev/md1) . it came back clean. so I ran fsck.ext3 -f and got lots of errors. Anyway, things are back to normal now. So today I installed etch on a new box to try and learn more about this. To test, I start copying onto the mdadm device and then hit the reset button on the pc. On boot, everything starts as normal. no resync, nothing. Can someone shed some light on this for me please? Does it have something to do with initramfs? Thanks. Chris. - To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
