[chiglug] Testing SAS drives

[Lincoln Bryant]

Hi,

It sounds like you could do a bunch of testing while the drives are still online and visible in GNU/Linux. I find the most obvious indicator of badness is things like reallocated sector counts on disks. Usually you can get smartmontools to give you information on individual disks behind a controller with something like:

$ smartctl -a -d megaraid,N
 where "N" is the number of your disk (0-15 i assume in your situation)

Wikipedia, naturally, has a list of attributes that can indicate failing-ness: https://en.wikipedia.org/wiki/S.M.A.R.T.#Known_ATA_S.M.A.R.T._attributes

smartctl will also give you the serial number of the disk, so you could correlate that with the disks as you pull them out.

Not a silver bullet, but a way to weed out obviously bad disks. I essentially throw out anything that has a reallocated sector count of >0.

-L
________________________________
From: discuss <discuss-bounces at lists.chicagolug.org> on behalf of L. V. Lammert <lvl at omnitec.net>
Sent: Friday, May 1, 2020 5:03 PM
To: discuss at lists.chicagolug.org <discuss at lists.chicagolug.org>
Subject: [chiglug] Testing SAS drives

Getting ready to swap the 2T drives in an MD120 drawer for 3T drives, ..
and I hate the idea of trashing 16 2T drives when almost of them are still
good.

Having a heck of a time figuring out any option to hang an SAS drive on a
PCI/PCI3 slot so the drive can get tested, as I have not seen any sort of
test option on the RAID controller.

Any options?

        Thanks!

        Lee

_______________________________________________
discuss mailing list
discuss at lists.chicagolug.org
https://lists.chicagolug.org/mailman/listinfo/discuss
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.chicagolug.org/pipermail/discuss/attachments/20200501/57c84c99/attachment.htm>