[Last Call] Learn how to a build a cloud-first strategyRegister Now

x
  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 3308
  • Last Modified:

Dell PowerEdge 4400 RAID errors

I am running Windows Server 2000 SP4 on a Dell PowerEdge 4400 server console. RAID 5 is in place and I have been getting some errors lately that have been scaring me. I looked them up on the EVENTID website and couldn't find anything useful. Went to the Dell forums site and couldn't find anything useful. Can someone please help me out here and give me some suggestions as well. Starting this off at 400 points.

Here are the errors:

1)      Source: AFAmgt  EventID: 20
\Device\AFA0 : ID(0:02:0); Error Event [command:0x28]  

2)      Source: AFAmgt  EventID: 20
\Device\AFA0 : ID(0:02:0); Unit Attention [k:0x6,c:0x29,q:0x3]  

3)      Source: AFAmgt  EventID:  20
\Device\AFA0 : ID(0:02:0) Timeout detected on cmd[0x28]  

4)      Source: AFAmgt  EventID:  20
\Device\AFA0 : SCSI Channel[0]: Timeout Detected On 1 Command(s)  

5)      Source: AFAmgt  EventID:  20
\Device\AFA0 : ID(0:02:0); Aborted Command [command:0x28]
0
Trihimbulus
Asked:
Trihimbulus
  • 4
  • 3
  • 2
2 Solutions
 
rindiCommented:
This looks like a problem with one of your disks. First check if there is no problem with the connectors of this disk, maybe the power connector is bad and intermittant. If maintenance on the contacts doesn't help, it is time to replace that disk. It looks as if this is one disk that has problems and the rest seem OK.
0
 
TrihimbulusAuthor Commented:
Let me ask you this, in the event that the hard drive needs to be replaced, how do I go about replacing and configuring the new hard drive. I have never done this before. The server has an application named FAST (flexible array storage tool). I am guessing this is where the configuration would take place. Are you familiar with this software?
0
 
rindiCommented:
This is hardware raid, I take it? Just get a disk of the same type and size, and replace it. The rest should then be automatic. The array will be rebuilt with the new drive (This will take some time during which the system will seem somewhat sluggish, this depends on the number of disks and their sizes). The Dell Utility software will help you monitor the state of the array.
0
What does it mean to be "Always On"?

Is your cloud always on? With an Always On cloud you won't have to worry about downtime for maintenance or software application code updates, ensuring that your bottom line isn't affected.

 
Duncan MeyersCommented:
The errors indicate that Disc 2 is crook. If you open the bezel's front door you'll see that the disc numbers are printed on the front panel. Disc 2 on the front is, in fact, disc 2 (surprise!). Once you have a new disc, whip out the old one (the front panel led will turn orange), replace the disc in the carrier (you'll need a phillips-head screwdriver and an anti-static mat)  and whack in the new one. The array will rebuild automatically. If it doesn't (unlikely) there is a BIOS setting taht needs changing. Easy to do - post here if necessary.

BTW - if you have Dell's extended warranty, call them! They'll make yoiu jump through a few hoops (like upgrading drivers and firmware first), but they'll organise you a new drive. Given the age of the system, your system may be out of warranty/maintenance.

Once the rebuild is complete (figure on an hour per 18GB), you may want to consider upgarding drivers and firmware of the RAID controller. You'll most likely have a PERC 2 controller (at an educated guess) - certainly one of the Adaptec range. You need to update the drivers first, then update the controller firmware from (again, an educated guess) 2.5 to 2.6 to 2.8 (major revisions). It's not as fearsome as it sounds - just follow the instructions on the website, or post here for more info. Most important thing: Upgrade the drivers **BEFORE** upgrading the firmware. Otherwise the server won't boot.

http://support.dell.com/support/downloads/devices.aspx?c=us&cs=04&l=en&s=bsd&SystemID=PWE_PNT_XEO_4400&category=0&os=WNT5&osl=EN

Once you've upgraded, I'd suggest getting rid of FAST and installing Array Manager. Scroll down to Systems Management at the link above.
0
 
TrihimbulusAuthor Commented:
Thanks so much!! I will post back in a few days- the HD is being ordered today!
0
 
TrihimbulusAuthor Commented:
Ok well we have ended up ordering 3 brand new 73Gb SCSI hard drives. The old hard drives are 33.8 Gb's. Is it possible to migrate the data from the old hard drives onto these new ones?? How in the heck do I do this one????
0
 
rindiCommented:
First make sure you have working, good backup as usual (test it as usual, too).

How many Disk slots does your server have? What raid controller model and revision, including firmware version is installed?

If your raid controller supports this (normal raid controllers used by compaq do support it, but I've been told several of Dell's Perc Controllers don't), you can expand a current array to a larger size by using the utility software, this means if your array has been added disk space, there is an "expand" option, which will allow you to enlarge the array to the maximum allowed by the the used disks. If you controller has this possibility, first replace the bad disk, then wait until the array has been completely rebuilt. Your array utility software should tell you when that is the case. Then replace the next disk and again wait until the array has been completely rebuilt. Now again wait for the rebuilding to finish until the you replace the last drive. Again after having done the rebuild, use the utility software to "expand" the array to use the full new diskspace. After that you caneither create an additional partition using diskmanager, or you can try to extend the existing partition using the commandline utility "diskpart", which has an "extend" option. You can try using this option to change the current partition sizes of your array. This doesn't allways work, if it doesn't it will tell you that that operation isn't possible for your config, so unless you make a big mistake, you can't really wreck anything yet. Use diskpart /? to get a list of commands and the syntax. If server 2k doesn't yet have diskpart included, copy a version from a 2k3 server, or even from an XP box. They should work on your OS too. If diskpart won't let you extend your current partitions, you could also use some 3rd party utility like acronis disk director suite to manipulate the partition sizes.

Ïf your PERC controller doesn't support expanding, but you have enough disk slots available, you could setup a new array with new disks, while the old array is still running, the you could try cloning your first array to the new one using a programm like acronis trueimage server (http://acronis.com). Once the clone to the new array has finished, you should be able to use your new disks as before the old ones...
0
 
TrihimbulusAuthor Commented:
1) Server has 6 slots (3 full)
2) Driver Provider is: Adaptec
    Driver date is: 05/03/2002
    Driver version is: 2.7.1.4944

    Software INfo:
    FAST version: 2.1 - 3 (Build#2941)
    Host driver version: 2.1 - 3 (Build#2941)
    Hardware INfo:
    Controller Type: PERC 3/Di
    Controller version: Does not apply
    Physicall PCI Info: Bus:0, Device:1
    BIOS version: 2.7 - 1 (Build#3571)
    Kernel version: 2.7 - 1 (Build#3571)
    SCSI Support: 1 Channel, up to 15 devices

Do I need to ditch FAST and get Array Manager???
0
 
Duncan MeyersCommented:
I would definitley recommend going to Array Manager, but you'll need to update driver (*always* upgrade the driver first for Adaptec RAID controllers)
and firmware to 2.8... (link as per above post)

>Once the clone to the new array has finished, you should be able to use your new disks as before the old ones...
Unfortunatley, not with a PERC 3/Di... It will *always* want to see 2 sets of three discs and so will always be in an error state.. You cannot fix the config like you can with an AMI controller. I'd recommend not going down that path.

The answers to your other question: http://www.experts-exchange.com/Storage/Q_21430260.html give your best next steps...

0

Featured Post

Get your Disaster Recovery as a Service basics

Disaster Recovery as a Service is one go-to solution that revolutionizes DR planning. Implementing DRaaS could be an efficient process, easily accessible to non-DR experts. Learn about monitoring, testing, executing failovers and failbacks to ensure a "healthy" DR environment.

  • 4
  • 3
  • 2
Tackle projects and never again get stuck behind a technical roadblock.
Join Now