I recently picked up one of those super micro 4u servers from Tams and swapped everything over but couldn't get the network to work. I finally got that fixed but then I started having kernel panics. Currently I"m stuck at "Kernel panic - not syncing" <4>REISERFS panic (device md2): journal-003 check journal_end: j_start (2835) is too high

 

This is after I was having problems with md1 so I ran a reiserfsck that told me to rebuild tree so I did. When the system was coming up disk 3 was showing as a red ball although fdisk -l confirmed all my disks were there. I took the array offline to go into maintenance mode and then it showed Parity drive was not there even though when the array was up it had a green ball. After running reiserfsck rebuild tree on md1 it appears that parity is ok in unraid now but when I reran the check on it all sorts of stuff was coming up although I didn't get a chance to finish the check because I need to shut down for this morning for a bit and it was taking too long.

 

Anyway, When I went to take it out of maintenance mode so I could capture syslog because /boot wasn't writeable that is when I had the above kernel panic on md2. So I'm not really sure where to go. Should I just put all these drives back in the old case since everything was working fine then and then try to get back to a working system... or what? I don't really know where to go from here.

Attached is the telnet output for the reiserfsck on md1 from last night

reiserfsckmd1_rebuild_tree.txt

I should also note I'm on 5.0-rc12a. Also when I tried to shut down my array so I could put it into maintenance mode last night it hung but I think since md1 was read only already because of errors it let me go ahead with the reiserfsck. This morning I didn't have any problems getting into maintenance mode after I rebooted.

So I just found this and now I'm hoping I didn't hose my system somehow. http://lime-technology.com/wiki/index.php/Check_Disk_Filesystems

 

I still don't really know how to proceed but if I don't here anything on what to do by tonight I"ll report back the steps I'm taking for posterity.

I came home to find out that disk 1 is showing unformatted now while disk 3 is still red balled. Which I'm sure is because I ran reiserfsck --rebuild-tree and also probably because when I did that I wasn't in maintenance mode. Had I been in maintenance mode parity would have picked up the changes. So I'm an idiot.

 

Then I realized my flash drive was corrupt which is why I couldn't write the syslog to it. I've taken care of that thanks to a tip from the amazing Joe L in another thread to run chkdsk on it from windows. That looks good so here is my most recent syslog attached.

 

I still don't know where to go from here. I think I will first try to run some smart commands on my disks to make sure they're good. If they are I'll probably start the reiserfsck on md1 again.

syslog-20130530.txt

Now my entire daggone flash drive was corrupt and chkdsk repair moved every file to lost and found except the 2 smart reports I grabbed before shutting down last time. Holy cow. Now I have no idea what is going to happen. Based on this here thread I should be able to just reinstall onto my flash drive, drop my key in, and go. http://lime-technology.com/forum/index.php?topic=20505.15

 

So now I'm resyncing parity. Again, attached are the 2 smart reports. SDB was red balled and SDC is currently showing unformatted. On a positive note sdb no longer has a red ball next to it and I'm pretty sure I don't see any issues with it in the smartlog.

smart_sdb.txt

smart_sdc.txt