Start a Conversation

This post is more than 5 years old

Solved!

Go to Solution

31860

July 30th, 2013 12:00

T605 Perc 6/I failed drive, replace with different capacity/brand

PE T605 with Perc 6/I three drives, Raid 5. One drive has failed, they are Seagate 250gb SATA 7200rpm drives. I could order one of these drives and wait a day or two or can get a different capacity/brand today and replace it tonight. The replacement is a Western Digital 1TB 7200rpm "enterprise" drive. I just want to verify that this will work, I realize it will only use 250gb of the 1TB so I'm losing that capacity but are there any other specs I need to look for or verify? or is it best to wait for an identical drive?

Also, the existing drives apparently need a firmware update re: http://www.dell.com/support/troubleshooting/us/en/04/KCS/KcsArticles/ArticleView?c=us&l=en&s=bsd&docid=DSN_619147E926299297E040AE0AB8E103AE this indicates that the MA07 needs to be upgraded to MA08. There is no MA08 just MA10 so I will install that. I've not done a firmware update on a drive, will this require a restart of the server? The controller firmware is also out of date, same question with this will it require a server restart? I plan to do the firmware upgrade after the replacement drive is installed and the array is rebuilt and a backup run, that seems the most prudent plan, or is it better to do the firmware upgrades first?

Thanks!

 

7 Technologist

 • 

16.3K Posts

July 30th, 2013 13:00

I would wait ... an "identical" drive would be a certified drive for this controller and you would eliminate the chances of firmware incompatibilities.  I would especially recommend this if the server hosts critical data/services.  Introducing a non-certified drive (although an enterprise-class drive) could cause unpredictable issues down the road.

Most drive update utilities are bootable, so a restart would be necessary.  The PERC firmware also requires a reboot, and should be done while the array is healthy (after it rebuilds).  Be sure to update your LifeCycle Controller/iDRAC and BIOS before updating the PERC.

59 Posts

July 30th, 2013 14:00

Thanks. The BIOS is current. Which to do first - the drive and rebuild of course but after that - the controller first then drives? or does it make a difference. We don't use Microsoft System Center so is the LifeCycle controller update necessary?

7 Technologist

 • 

16.3K Posts

July 30th, 2013 14:00

Best to update LLC/BIOS before anything else, then peripheral components (RAID, NIC, etc.), then drives.  It shouldn't really matter, but there is the possibility that the drives will be updated with code/responses/etc. that the controller doesn't understand (yet).

The LLC has nothing to do with SCCM, so, yes, it is necessary.  On older systems, the ESM (Embedded Server Management) worked with the BIOS to manage/monitor the system hardware - problems could arise when they are not updated together (generally BIOS first).  LLC is a next-generation ESM and is increasingly becoming the main controller of a server and should be updated first, then the BIOS.

7 Technologist

 • 

16.3K Posts

July 30th, 2013 19:00

You can probably find it in the BIOS utility for it, but you can find it using OMSA (under System, Main System Chassis, Firmware) ... if you have that installed.

59 Posts

July 30th, 2013 19:00

How can I tell what version of LifecyleController and DRAC are currently installed?

59 Posts

August 1st, 2013 11:00

I do have OMSA installed. Main System Chassis, Firmware says Baseboard Management controller is version 2.30.00. The latest (only) LifeCycle controller in drivers and downloads is version 1.3. Under System Management there is the full LifeCycle package, listed as optional. I don't think this was ever installed.

 

7 Technologist

 • 

16.3K Posts

August 1st, 2013 12:00

Sorry ... I forgot we were talking about the T605.  It is classified as an "older" model and does not have the same LLC as the newer servers.  Update the BIOS, then ESM/BMC firmware, then PERC and other peripherals, including DRAC (if installed, which is the predecessor to the LLC/iDRAC).

7 Technologist

 • 

16.3K Posts

August 1st, 2013 20:00

59 Posts

August 1st, 2013 20:00

 

The bios is up to date, DRAC is not installed, I've downloaded the hard drive firmware update, the BMC update,  the Storport driver, the Perc 6/I 6.3.3-0002 firmware. However, for the life of me I can't find the Driver Version 2.24.00.32 I only see the 2.23 listed.

Service tag for this unit is [ADMIN NOTE: Service Tag removed per privacy policy]

I'll be replacing the failed driver tomorrow after hours so I will probably wait and do these updates after the rebuild and a full backup.

Thanks!

7 Technologist

 • 

16.3K Posts

August 4th, 2013 13:00

There is no reason to "clear" the drive if you have not yet assigned it as a hot-spare and it was not participating in any array.

59 Posts

August 4th, 2013 13:00

I replaced the drive on Friday, turns out this was not hot swappable as I thought so I did have to power down to replace the drive. I had bought 2 drives, one to act as hot spare. After installing the drives the server would not boot, got power but nothing at all on the monitor. I thought it was the server or drives, reseated them tried again, nothing. The monitor was flashing green which I thought was weird, felt like if there were a drive problem I should still get BIOS. Finally tried different monitor, that was it, whew! I was able to go into controller and rebuild the array. After system booted up and I made sure everything was working I left because it was quite late. Logging in remotely it is showing the extra drive as predictive failure, I had not yet assigned it as a hot spare. I'll take the faulty drive out but should I clear that drive? I also found that one of their redundant power supplies was dead, saw this in OMSA before but had thought someone had unplugged it. Once I get the replacement power supply I'll go out and take out the bad drive.  

No Events found!

Top