Dell Perc H710 predicted failure
-
I have a Dell server with a Perc H710 drive. Has three virtual disks. Raid 1 os, single disk temp drive, and 3 disk Raid5 (I know, bad, but I didn't set it up) The Raid 5 has a disk marked with predicted failure. I have two new replacement drives and wanted to run my process by an extra pair of eyes.
I have read to fail the bad drive using open manage then pull and insert new drive and allow a rebuild, however, is that the best route? Couldn't I put in a new drive, mark it as a hot spare, wait for that process to complete, then pull the bad drive? Wouldn't that be a faster more reliable way?
Thanks!
-
@i3
Make a backup of the System and then wait for the hard drive to fail or if it was Dell Support, what they will tell you to mark the drive as offline and then pull it and add the new drive and it will rebuild. You can also use a hotspare but it will add more problems for the rebuilding due to the nature of it. -
Mark the drive offline and swap it. Period.
-
Don’t wait don’t delay just fucking do it
-
@dbeato said in Dell Perc H710 predicted failure:
@i3
Make a backup of the System and then wait for the hard drive to fail or if it was Dell Support, what they will tell you to mark the drive as offline and then pull it and add the new drive and it will rebuild. You can also use a hotspare but it will add more problems for the rebuilding due to the nature of it.But I would just replace it...now
-
Like @JaredBusch already said, get the drive replaced now. Just make sure you have a good backup first.
-
Thank you for the replies. I do have a backup and I plan on replacing it ASAP. My question is best way to replace it.
Fail the drive and let it rebuild or put new drive in allow it to be assigned as hot spare then fail bad drive immediately.
My understanding is the first option would involve the array to be degraded and rebuilt taking significantly longer to rebuild and being vulnerable while rebuilding.
Wouldn’t the second option copy the data off just the drive in question then take the bad drive offline?
Currently the virtual disk is still healthy and the predicted failure disk is still working. Option two seems like less risk and a quicker repair.
-
@i3 said in Dell Perc H710 predicted failure:
array to be degraded and rebuilt taking significantly longer to rebuild and being vulnerable while rebuilding.
Wouldn’t the second option copy the data off just the drive in question then take the bad drive offline?A hotspare will do a rebuild still.
-
I told you already. I did not think I used too many words.
@jaredbusch said in Dell Perc H710 predicted failure:
Mark the drive offline and swap it. Period.
Nope, not too many words.
-
You have a PERC, it is perfectly capable of a flat out blind swap. Manually marking it is just for your own jollies.
-
@jaredbusch said in Dell Perc H710 predicted failure:
Mark the drive offline and swap it. Period.
This, don't get weird or complicated. Anything other than this just adds effort and risk.
-
@dbeato said in Dell Perc H710 predicted failure:
@i3 said in Dell Perc H710 predicted failure:
array to be degraded and rebuilt taking significantly longer to rebuild and being vulnerable while rebuilding.
Wouldn’t the second option copy the data off just the drive in question then take the bad drive offline?A hotspare will do a rebuild still.
Not always, hot spares don't always have the same clarity of action as a proper swap.
-
@i3 said in Dell Perc H710 predicted failure:
My understanding is the first option would involve the array to be degraded and rebuilt taking significantly longer to rebuild and being vulnerable while rebuilding.
This situation happens either way. Only the proper first method fixes it faster than the unnecessary second method. The hot spare method has no advantages, just isn't quite as clear and moves your drives to weird positions (which is fine, but why?)
-
@i3 said in Dell Perc H710 predicted failure:
Wouldn’t the second option copy the data off just the drive in question then take the bad drive offline?
No, that is not how a hot spare works. A hot spare is just sitting there doing nothing, waiting for a drive to fail.
-
@i3 said in Dell Perc H710 predicted failure:
Currently the virtual disk is still healthy and the predicted failure disk is still working. Option two seems like less risk and a quicker repair.
It just feels this way, but this is incorrect.
-
Thank you for all the replies. I will proceed with the mark failed and swap.
Thanks again!
-
Update: I had to clear the foreign config on the new drives, however, Open Manage would not display the "are you sure you want to clear the foreign config" message. It just went to a blank page. I had to uninstall OM and upgrade to the current version, then I was prompted about firmware and driver versions. To be safe I upgraded those as well, with reboots in between, took another system image failed the "bad drive" popped in a new one and the rebuild began. Completed in about 5 hours. All is working well.
Thank again all.
-
@jaredbusch said in Dell Perc H710 predicted failure:
Don’t wait don’t delay just fucking do it
You should start making tshirts with this quote
-
@brrabill said in Dell Perc H710 predicted failure:
@jaredbusch said in Dell Perc H710 predicted failure:
Don’t wait don’t delay just fucking do it
You should start making tshirts with this quote
Learn something today from your failure to reboot more did we?
-
@jaredbusch said in Dell Perc H710 predicted failure:
@brrabill said in Dell Perc H710 predicted failure:
@jaredbusch said in Dell Perc H710 predicted failure:
Don’t wait don’t delay just fucking do it
You should start making tshirts with this quote
Learn something today from your failure to reboot more did we?
Not really.
Wouldn't have helped. Would have just messed up whatever weekend i did it on.
