After running UnRaid without any problem for months, I just experienced my first crash - I got a new WD 1TB disc and the server crashed when it tried to clear the disc, about 15% into the procedure (I tried it several times with the same result). I have taken the drive out and tried it on my XP machine, which seems to work fine - I can format it, copy files to it, ran error checking and XP found no error. I have also tried the Preclear_disk.sh script but that caused the server to crash as well. My question is how can I capture the error in the syslog just before it crashes? After I reboot the server, the syslog file is reset and I cannot see what happened... Are there other utilities that I can run to see if the drive is defective (even if XP thinks it is ok)?

 

Thanks in advance,

Jaz

 

 

After running UnRaid without any problem for months, I just experienced my first crash - I got a new WD 1TB disc and the server crashed when it tried to clear the disc, about 15% into the procedure (I tried it several times with the same result). I have taken the drive out and tried it on my XP machine, which seems to work fine - I can format it, copy files to it, ran error checking and XP found no error. I have also tried the Preclear_disk.sh script but that caused the server to crash as well. My question is how can I capture the error in the syslog just before it crashes? After I reboot the server, the syslog file is reset and I cannot see what happened... Are there other utilities that I can run to see if the drive is defective (even if XP thinks it is ok)?

 

Thanks in advance,

Jaz

 

 

In almost every case, a crash is an indication of a hardware problem, and it is unlikely to be specific to a drive, but more likely to be an issue with memory, or the power supply... or heat

 

To try to capture the syslog, especially if the crash is repeatable, you could open up a telnet session to the server and type:

tail -f /var/log/syslog

 

Then, in a different telnet session you can start the pre-clear script. (easier to recover from than when adding the disk to the array and it clearing the drive)

 

You can run the smartctl program on the drive.  You can ask it to perform a long test on itself (usually takes a few hours before you can get a status report to see how it did)

 

Before you go through all that, run the memtest program in the boot menu.  If it is clean through several cycles, you can proceed.. If not, check its voltage, timing, and frequency settings for your specific memory strips.  It might be as simple as that.

 

Joe L.

Joe,

 

Thanks for the quick response, I will follow your instruction and find out what is causing the crash.

 

[update] I think the problem was temperature and/or power supply related, since I moved some drives out of the server (from 10 drives to 5 drives) into my new eSATA cabinet and now everything is running much cooler. Unless it acts up again, consider this one fixed! 

 

[update2] it has been another week of trouble-free operation, so I think I can safely conclude that the crash was caused by overheating of some sort.

 

Jaz