When I got home yesterday, my Tower was frozen either from a power surge or something else.  I gave it a hard reboot since it was non-responsive from the GUI.  Now, it's hitting an error when it tries to mount the disks and just hangs forever.  I was watching the Log the last time I tried to start up the server (attached) but here's the interesting part:

 

Oct 25 07:21:54 Tower kernel: XFS (sde1): Internal error XFS_WANT_CORRUPTED_GOTO at line 1627 of file fs/xfs/libxfs/xfs_alloc.c. Caller xfs_free_extent+0xbe/0xf1

Oct 25 07:21:54 Tower kernel: CPU: 0 PID: 4308 Comm: mount Not tainted 4.1.18-unRAID #1

 

Oct 25 07:21:54 Tower kernel: XFS (sde1): Failed to recover EFIs

Oct 25 07:21:54 Tower kernel: XFS (sde1): log mount finish failed

 

I'm completely out of my depth.  What do I do now?  Thanks so much for any help.

Log.txt

You need to start the array in maintenance mode and check filesystem on sde1 which I assume is your cache disk.

I started the server in Maintenance Mode and checked the filesystem status of the Cache drive with -nv.  The results are attached.  It ran the check really fast.  I feel like it should have taken longer.  I'm not sure how to make sense of the output as it doesn't call out any obvious errors (to my untrained eyes).

xfs_repair_status.txt

I ran the check again with -L and then the server was able to start up and mount disks correctly.  I guess something was corrupt on the Cache drive?

I started the server in Maintenance Mode and checked the filesystem status of the Cache drive with -nv.  The results are attached.  It ran the check really fast.  I feel like it should have taken longer.  I'm not sure how to make sense of the output as it doesn't call out any obvious errors (to my untrained eyes).

actually the output shows that there was a LOT of corruption!    Now that the disk appears to be in a good state you might want to try running xfs_repair again to see what the output looks like when no issues are found.

Ha, goes to show you what I know.  I check it again when I get home now that everything seems to be running well again.