Drive Array Device Failure - SCSI Port 1, drive ID 6 failed - REPLACE

[Valeed Mulla_1]

Error: Drive Array Device Failure - SCSI Port 1, Drive ID 6 Failed

My server Configuration is as follows:

ML350 G3 server
Raid 5
Total number of HDD = 4
Drive capacity each = 36.4 GB (U320 SCSI)
Array Controller = SLOT 2, PCI Smart Array 641 Controller

Drive failure at bay 2
Error shows drive failure at bay 6 , but there is no bay 6 on the server maximum bays are 5 (from 0 to 5)

We have replaced the faulty drive with a new hdd but still it says the drive is faulty but instead of showing the drive ay bay 2 it shows that the faulty drive is at bay 6, please advice how to resolve this problem.

find attached the array diagnostic report

regards

Valeed Mulla

[Ali]

Hi Valeed,

This is a case of SCSI ID changed. this rearly happens. I experienced the same issue long time ago. the reason for this issue is not identified this could be due to sometime movement of the hard drive, hotplug replacement of the failed harddrive while server is up or firmware or could be something else.

you have only 6 Bays in the server where SCSI IDs 0-5 are fixed on the Hard drive backplane. You don't have Bay 7 SCSI ID 6 in this server so you can't replace the hard drive.

Only workaround to this problem is to take the complete back as your raid is still there on degraded state.

remove the existing raid setup, update system bios and smart array controller firmware if not latest.
Create new array and logical drive and restore the backup.

Hope this helps,
thanks,
Aftab

[Ali]

Hi Valeed,

please don't forget to assign points if you think information provided in previous post is is helpful... :)

Cheers!

Aftab

[Valeed Mulla_1]

Dear Aftab,

Thanks for your reply, but this is the last resort that i can do , since this is production server and there is a business application running on the server.

To perform your suggestion I have to first have a fall back solution - i have a spare server on which i want to restore the backup of the production server, I made a full ASR backup of our production server but bad luck that the asr fails to recover after one hour with the error code.

The recovery application â ntbackup recover /1 /sifpath=C:\windows\repari\asr.sifâ returned an error code 0x2. Since this indicates an unrecoverable error, asr cannot continue on this machine.

I have recovered earlier through ASR tapes on the same server and it use to restore correctly, but dont know why its failing to restore this time, I have even tried with three different ASR backup sets which was made in september 2010, April 2010 and even last years ASR set made in April 2009 which was a less capacity - but all of them fail with the same error - as mentioned above.

can you please advice what makes it fail , is there anything else that we should do in order to have a failsafe recovery.


Please reply URGENT - this is a critical situation.

Thanks for your support

regards

Valeed

[Ali]

Hi Valeed,

Before performing an ASR restore, ensure that the target system to which the restore will be made meets the following requirements:

The target system hardware (except for hard disks, video cards, and network adapters) is identical to that of the original system.


There are enough disks to restore all the critical system disks.


The number and storage capacity of the critical disks are at least as great as those of the corresponding original disks.

Let me understand your backup setup
what is the back drive / tape drive that you are using ?

Are you storing ASR Backup on the same server or on identical server.. ? same server may not work as it is runnding on raid degraded state. Raid 5 on interirm recovery mode (degraded) is extremely slow interms of I/O operations.

Many HP Tape drive comes with OBDR (one button disaster recovery) which is much better and safer backup / restore than NTbackup ASR.

So explain your backup setup.

thanks,
Aftab

[Valeed Mulla_1]

Dear Aftab,
thanks for your reply, I have been trying to do all teh possibilities to check and see whats going wrong and where, as advised i have done all teh checks as listed below:

we have 2 servers production and test both are identical except test has only 1 gb ram where as production has 3 gb ram.

both have same HDD - RAID5, 4 disks 36.4gb each

Tape drive on both servers HP DLT 40/80 , this tape drive is internal and doesnt have the OBDR switch

ASR is restored on test server

I have backup sets - daily , weekly, monthly and - monthly ASR backup sets also.

I have tried to restore through ASR - tried around 5 tape sets ranging from this year to last year but all of them fail with the same error.

The recovery application â ntbackup recover /1 /sifpath=C:\windows\repari\asr.sifâ returned an error code 0x2. Since this indicates an unrecoverable error, asr cannot continue on this machine.

then i tried to restore the active directory / system state by using normal backup tape - but that too fails with the below error:

Lsass.exe â system error , directory services could not start because of following error: The system cannot find the file specified. Error Status:0xc000000f.Please click ok to shutdown this system & reboot into directory services restore mode, check the event log for more detailed information.


tests performed to check hardware:
chkdsk - status ok
swapped tape drive - status ok
installed win2003 std ed sp2 -- status ok
formatted logical drives - NNTFS (not quick low level) -- all ok

After doing all the above activity and still failing to restore from backup , I beleive itâ s not hardware but a software issue thatâ s stopping the server from restoring, appreciate if you can look into the matter and advice what can be done next in the row of steps to perfomr teh ASR suecessfully.

regards

Valeed

[Michael A. McKenney]

You need to install the OS and do a system state restore of the AD. Do you have another AD server you could use to rebuild AD on that server.

Microsoft does not support RAID 5 for the OS. Only RAID 1 and RAID 10. The paging file should be in RAID 1 and RAID 10.

Unrecoverable errors are probably a corrupted tape. Did you try cleaning your tape drive 5x. Do you have another tape to recover from?

[Ali]

Hi Valeed,

I will explain this in details later on.

In brief what I understand that the target system does not meet the h/w requirement for ASR restore. you mentioned that you have just 1 GB memory.

Take a normal Full backup including system state from the source server. ASR backup will fail if your hardware is not identical but normal full backup can be restored with few errors.

Check the amount of memory resources used, Virtual memory size allocation on the source server and try to make the same settings on your target system before restoring the backup.

Also take small backup (probably just bunch of folders) and restore them to validate your tape, drive and backup software.

Hope you have installed a fresh copy of same windows on the target server what you have on the source server before restoring backup.

thanks,
Aftab

[Michael A. McKenney]

Another method I use in Windows is from command prompt using Xcopy. Windows copy overhead causes issues.

Xcopy source target /s /e /v /y

Xcopy is faster than Windows copy. You also want to disable anti-virus and make sure all applications are disabled in services and no users are accessing the files to prevent open files.

[Valeed Mulla_1]

Dear Aftab,
Thanks for your reply, both servers are identical except memory source has 3gb test has 1gb only, I have tried to restore the AD through "Directory Services in restore mode (Domain controllers only)" but that also fails with the below message:

Lsass.exe â system error , directory services could not start because of following error: The system cannot find the file specified. Error Status:0xc000000f.Please click ok to shutdown this system & reboot into directory services restore mode, check the event log for more detailed information.

I had a suspect that the tape drive could be bad , to check on this i have swapped the source server tape drive to test server - also cleaned it with cleaning cartridge before performing restores - but still it fails

regarding virtual memory size allocation if i am doing the ASR teh system formats all data and then performs - so it will clean the memory settings as well - does it really make a difference ? because i had done the restore earlier with the same amouunt of memory and it worked.
please advice.

regards

Valeed.