Unsolved
3 Posts
0
1002
Raid5 Drive failed, another in prevented failed state
I've a old Poweredge r710 that was inherited four years ago after some years in a company. It has four hdd seagate 1TB each configured in Raid5. One month ago one of the drives enter in prevented fail state (light blinks amber and green). Some days after another drive enter in prevented fail state and subsequently failed after two days (I read the log and the drive failed by a lack of communication: command timeout 2094).
I know that this si an aged server but I was wondering if it could go ahead with it for a couple of years, as the task that is accomplishing is not very challenging. But I don't want to spend money in a aged machine, so I just bought an used Seagate from the same model (ST31000424ss) to try to replace the failed drive.
My question is: what steps should I take now? I don't know how is configured the used disk that I bought or where have it been used before. I assume it is in a good state.
Should I just replace directly the failed drive with the used one and expect that it will rebuilt it? Do I need to prepare the drive before the replacing act? And if yes: how? I've installed OSA on the server.
All the help will be welcome!
DELL-Chris H
Moderator
Moderator
•
8.4K Posts
0
June 5th, 2020 13:00
Arealvel,
First off make sure you have a complete backup.
The timeout is a sign the controller and drives are out of date on firmware. I would start by walking the BIOS and iDrac up to date together, followed by the raid controller and drives. After that then try rebuilding the failed drive, and run a Consistency Check on the Virtual Disk and let me know what you see.
arealvel
3 Posts
0
June 6th, 2020 02:00
Thank you for your quick answer.
My actual versions are:
BIOS: 6.4.0
Idrac6: 1.96
DLC: 1.6.0.73
Perc6i integrated (Embedded): 6.3.3.002
So I should begin with the Bios and the idrac6 update. Here I found:
last version of Bios is 6,60 (File for WS2012R2 is BIOS_0F4YY_WN64_6.6.0_01.EXE)
last version of idrac6 is 2.92. (File: ESM_Firmware_KPCCC_WN32_2.92_A00.EXE)?
May I update directly to this versions?
and
As they are exe files, may I directly run them locally from the OS? or should I use other way?
DELL-Stefan R
Moderator
Moderator
•
790 Posts
0
June 8th, 2020 03:00
Hi,
well, you might go for the SUU (Server Update Utility). Found them here:
DELL EMC Server Update Utility, Windows 32 bit Format, v.17.12.00
https://www.dell.com/support/home/en-us/drivers/driversdetails?driverid=twh73&productcode=poweredge-r710
DELL EMC Server Update Utility, Linux 32 bit Format, v.17.10.00
https://www.dell.com/support/home/en-us/drivers/driversdetails?driverid=9ghnc&productcode=poweredge-r710
Regarding your question, on old systems, I recommend walking the updates step-by-step instead of installing the latest directly. You may run into issues with the updates.
And, yes, you can run them directly on the operating system.
Best regards,
Stefan
arealvel
3 Posts
0
June 13th, 2020 07:00
Hi, Stefan. Thank you for the links and for the advices. I was a little scared about the size of the ISO file, so at least I optet to update manually the main components of the PE R710: Idrac6 DLC, Bios, Raid Controller (For interested people to update the outdated Idrac6 there is some relevant information here. Useful also this (and this for DLC).
Now I am at the latest versions (BIOS 6.6.0 | IDRAC6 2.92 | DLC 1.7.5.4 (but this is just to avoid problems, as by now DLC has no any catalogue) | PERC 6/i 6.3.3.002 | Also the drives are updated.
After this I reseat the failed disk (disk 0) and PE began and completed the rebuilding process. The only problem was that during the first minutes of the rebuilding process the disk enters in a prevented fail state.
After that, I run a consistency check, which passed. But looking into the alert log I saw a bad block removed. That is perhaps the cause of the prevented fail state for disk 0.
I attach two pictures.
By now all is working. Thanks both of you for the help. If it'll fail I'll repurpose the initial question. Thank you again.