Hello All,

 

I was at work this morning and remoted in to check my server stats from the over night when i found that the array had both my parity drives listed as failed. Its been running fine for 17 days, after i did a h/w upgrade of motherboard, memory & controller. I began a parity check immediately and since then its been running for 13hrs now. The strange part is that at 50% it began running around 174 KB/sec (est 199 days 20hrs to finish). Below is the syslog of the presumed failure time-stamp.

 

 
Dec 16 02:32:51 Gibson kernel: mdcmd (397): spindown 3
Dec 16 02:42:06 Gibson kernel: mdcmd (398): spindown 1
Dec 16 02:52:38 Gibson kernel: mdcmd (399): spindown 9
Dec 16 03:23:14 Gibson shfs/user: err: shfs_rmdir: rmdir: /mnt/cache/auto-downloads/deluge/Incomplete (39) Directory not empty
Dec 16 03:34:02 Gibson shfs/user: err: shfs_rmdir: rmdir: /mnt/cache/auto-downloads/deluge/Incomplete (39) Directory not empty
Dec 16 04:40:01 Gibson rsyslogd: [origin software="rsyslogd" swVersion="8.16.0" x-pid="1716" x-info="http://www.rsyslog.com"] rsyslogd was HUPed

Fatal error: Allowed memory size of 134217728 bytes exhausted (tried to allocate 130752512 bytes) in /usr/local/emhttp/plugins/dynamix/include/DefaultPageLayout.php(300) : eval()'d code on line 73

 

Its mentioning something about the dynamix, but i'm unclear on what particularly? I have Dynamix S3 sleep, Dynamis System Autofan & Dynamix System Stats apps installed. Auto fan & System stats have worked from the before the mobo, mem & controller upgrade, but S3 sleep is relatively recent.

 

Both my parity drives are WDC_WD60EFRX (WD 6TB Red's). So assuming that the 2nd drive has had some sort of catastrophic failure. How "safe" am i to shut the system off, replace with a brand new drive and re-run parity check?

 

I've been trying to learn more about rsync for the very purpose of backing my primary data to my alternate unraid box, but alas timing is not on my side. My "belief" is that my primary data is intact and will be operational once the party check is completed, that only the "recent" data will be lost... but not data that is already stored on drives. Please let me know if there is anything else i can try/attempt with any amount of success.

 

 

Thanks,

 

 

Diagnostics?

Because of the error, it might be impossible to run diagnostics through the GUI.  I'd guess that you were trying to hit Tools - System Log.  There's an issue with HUGE syslogs and that System Log button.

 

Try

diagnostics

or

cp /var/log/syslog /boot/syslog.txt

 

Hello Trurl & Squid,

 

The actual syslog.txt resulted in 0kb, but it turns out there was also a syslog1 & syslog2. syslog1 is 127,680KBs. So I've looked through syslog1 and attached a .txt of some of the error messages.

 

 

Now i'm left wondering if its the new controller or the drive ? I moved from a Supermicro X10SLL+-F, Xeon 1230v3, 32GB ECC & IBM M1015 (flashed to IT)  to Supermicro - X8DT3, CPU E5620, 48 GB ECC & 2x AOC-S2308L-L8E.  Both setups use an Intel SAS expander.

 

I'm thinking the S2308L with the expander is too much for it or its glitchy somehow? Unfortunately I had to swap boards because the M1015 would not work in the X8DT3 board, it kept trying to boot from the card and not the USB (removing the sas cables from the M1015 did allow it to load, but with no drives ><).

 

Any thoughs on if I can make this work as is, before I dissemble two servers to swap boards and controllers?

 

Thanks again for any assistance,

syslog1.txt

Please just post the complete diagnostics zip.