Disk failed, now have a major problem I need help with

[acemilo]

Hey guys, I have 6 data drives in my server, one of which recently failed on me. I RMA'd the drive and got a replacement, slapped it in my server, and did a greyhole --fsck, which took about 5 days to complete since i have several TBs of data. When it was working, I noticed that anytime it wrote to the drive (/var/hda/files/drives/drive5) I could no longer access any of my shares either on the network or from the box itself (I have my shares mount locally to the server as well). Doing an ls on the /mnt/samba dir would just hang up on me and I'd get "resource not available" errors. Restarting the box would make everything work tho until greyhole starting copying stuff around again. Now that the fsck is finished, it was working great, until the daily fsck starts. Now it works fine all day, but when I get up in the morning, it must fsck overnight, because every morning for the past few days my shares are unavailable again until I restart my server. Obviously it shouldn't be doing this. What can I do to fix this? Restarting my server every morning isn't really a solution, it's just a band aid.

Thanks for the help

[sandstroem]

Hi,

Did you tell greyhole that your drive was gone for good?

Code: Select all

-g, --gone[=path] Tell Greyhole that the missing drive at <path> is gone for good. Greyhole will start replacing the missing file copies instantly. If you don't supply an option, the available options (paths) will be displayed.

Not sure if this is needed. However, when you replaced your drive with a new one the new drive will get a new uniqe UUID which you need to update in /etc/fstab

After, execute 'greyhole --fsck' and wait.