Koozali.org: home of the SME Server

Software Raid Array constantly rebuilding

Offline sits

  • ***
  • 68
  • +0/-0
Software Raid Array constantly rebuilding
« on: October 05, 2005, 07:15:32 AM »
I have a SME 6.01-01 server that is contantly rebuilding, one drive feels excessively hot, the log report says both drives are available, but the server will shutdown after a couple of hours, when its restarted the Array rebuilds. I havn't been able to determine cause and effect, is the raid rebuilding because of the crash, or would the crash be because of the heat generated from the drive under constant stress from the raid rebuilding (catch 22)

Any help would be appreciated

going around in circles here.
...

Offline smeghead

  • *
  • 563
  • +0/-0
Software Raid Array constantly rebuilding
« Reply #1 on: October 05, 2005, 07:38:55 AM »
.. try rebuilding with a fan aimed at the drive to reduce its heat stress .. if the process completes you have your culprit
..................

Offline sits

  • ***
  • 68
  • +0/-0
Software Raid Array constantly rebuilding
« Reply #2 on: October 05, 2005, 09:23:35 AM »
I have already added another 6" fan right on the drive housing still no change. in process of building another server to subsitute so i may bring it back to office and run diagnostics on it. heat may have affected something else. will let u know what i find out.
...

Offline Reinhold

  • *
  • 517
  • +0/-0
    • http://127.0.0.1
Software Raid Array constantly rebuilding
« Reply #3 on: October 07, 2005, 05:42:37 PM »
...try looking for an rpm for hddtemp
Note: hddtemp-0.3-0.fdr.0.10.beta10.rh80.i386.rpm
...will still work on standard SME 6.x...

install the rpm and (on the console)
hddtemp /dev/hd[abcd]

will (hopefully) tell "which if any" is too hot.

Regards
Reinhold
............

Offline sits

  • ***
  • 68
  • +0/-0
Software Raid Array constantly rebuilding
« Reply #4 on: October 08, 2005, 05:26:27 AM »
Thanks Reinhold,

That RPM is definitely needed I was able to stress test the raid and monitor the temp of the drives and hda was about 4 degrees hotter than hdc the mean temp was going to 52 degrees on /dev/hda outside the safe operating range for the seagate drives, causing a reboot, then the raid array was being rebuilt once again making the temp go over, (hence catch 22)

I have replaced the suspect drive added another cooling fan to the case, and this seems to have solved the issue. With both cooling fans operating mean temp under load is around the 42 degree mark

Thanks for your help
...

Offline Reinhold

  • *
  • 517
  • +0/-0
    • http://127.0.0.1
Software Raid Array constantly rebuilding
« Reply #5 on: October 08, 2005, 12:53:03 PM »
Quote from: "sits"
...mean temp under load is around the 42 degree mark


"Cool"
- pun intended  
:-D

Regards
Reinhold

P.S.: Don't forget about the Seagate 5year warranty.
May well be you'll receive a new drive. Just check on their website.
............