Hi everyone,

I was performing my monthly parity check (without auto-correction) on my unRAID 4.7. server. At some point it seemed to get stuck (i could not reach main or unmenu). Since i had cache dirs running on the background I thought that it might have something to do with the memory and i performed a hard restart, having removed cache_dirs from the go script.

 

The server restarts ok, i can access it locally and through ssh but main and unmenu do not start (first time i get a 'wait..' in my browser and when I refresh it seems to try to load indefinitely). I also cannot see the server, the shares or the flash  from my pc. I can also hear the HDs which probably means there is a parity check being performed after the improper restart, which is to be expected.

 

Can you please give me some advice? Thanks a lot...

syslog_00-32_14-1-2013.txt

What do you see on the system console?

What do you see on the system console?

 

I see what I think I am supposed to see. the login prompt. and i can login normally.

 

PS: i attached the syslog. i think i see something irregular (and a bit scary). Those 2 messages repeating a lot:

 

Jan 14 00:14:50 Tower kernel: handle_stripe read error: 1000/6, count: 1
Jan 14 00:14:50 Tower kernel: md: disk6 read error

 

 

Jan 14 00:15:21 Tower kernel: sas: --- Exit sas_scsi_recover_host
Jan 14 00:15:52 Tower kernel: sas: command 0xf2069240, task 0xf29cd640, timed out: BLK_EH_NOT_HANDLED
Jan 14 00:15:52 Tower kernel: sas: Enter sas_scsi_recover_host
Jan 14 00:15:52 Tower kernel: sas: trying to find task 0xf29cd640
Jan 14 00:15:52 Tower kernel: sas: sas_scsi_find_task: aborting task 0xf29cd640
Jan 14 00:15:52 Tower kernel: /usr/src/sas/trunk/mvsas_tgt/mv_sas.c 1701:mvs_abort_task:rc= 5
Jan 14 00:15:52 Tower kernel: sas: sas_scsi_find_task: querying task 0xf29cd640
Jan 14 00:15:52 Tower kernel: /usr/src/sas/trunk/mvsas_tgt/mv_sas.c 1645:mvs_query_task:rc= 5
Jan 14 00:15:52 Tower kernel: sas: sas_scsi_find_task: task 0xf29cd640 failed to abort
Jan 14 00:15:52 Tower kernel: sas: task 0xf29cd640 is not at LU: I_T recover
Jan 14 00:15:52 Tower kernel: sas: I_T nexus reset for dev 0400000000000000
Jan 14 00:15:52 Tower kernel: sas: I_T 0400000000000000 recovered
Jan 14 00:15:52 Tower kernel: sas: --- Exit sas_scsi_recover_host

At my 4th-5th attempt to restart, everything started fine.

 

I really have no clue why that happened, I changed nothing  and now I am wondering if I should do the parity check in case the same thing happens again....

Does disk6 have a non-zero error count on unRAID main?

Does disk6 have a non-zero error count on unRAID main?

 

No, disk 6 shows no errors. Unfortunately, the parity check showed 866 errors, but I am thinking they might be due to the repeated hard restarts

 

(Sorry for taking so long to respond while you try to help, too much work this week)