LSI Logic raid memory failure

[kemo]

hello

A repeated message coming on the screen telling below


Ext3-fs error ( device dm-2 ) in start-transaction: journal has aborted
Ext3-fs error ( device dm-2 ) in start-transaction: journal has aborted
Ext3-fs error ( device dm-2 ) in start-transaction: journal has aborted

Mega-raid cmm : controller cannot accept cmds due to the earlier errors


After reboot :


PwerEdge Epandable RAID Controller BIOS H418 (Build Key 83, 2004 )

Copyrig----- LSI logic corp

Press to configure

Memory/battery Problems were detected.

The adapter has recovered, but cashed data was lost.
Press any key to continue.

If you press it will HANG.

even if i go through the Ctrl+M to configure . it hangs.

Any clue?

Note:
A- we have 3 disks
2 disks as raid 1
1 disk standalone.
B- Server is DELL poweredge 2850
C- RHEL 4

Another query :
i have a scsi controller in the PCI with empty slots. any way to use it to boot the system UP even with single hard disk and no raid.

thanks

[Matti_Kurkela]

> Ext3-fs error ( device dm-2 ) in start-transaction: journal has aborted

This means the filesystem code has detected an I/O error and has transitioned to read-only mode. This might mean there has been a disk error and the RAID controller could not recover it.

> Mega-raid cmm : controller cannot accept cmds due to the earlier errors

Uh Oh. I'm not very familiar with the megaraid controllers, but this looks like the controller had locked up. You might have a RAID controller failure, which may be bad news.

> Memory/battery Problems were detected.

There has been a problem with the battery-backed write cache on the RAID controller. This means some data has been lost, and your filesystems will require a filesystem check at the very least. If this is the only problem, it is bad enough, but not as bad as the earlier messages suggested it might be.

> If you press it will HANG.

> even if i go through the Ctrl+M to configure . it hangs.

This suggests a possible RAID controller hardware problem that prevents the system from even trying to boot.

It might be necessary to move the disks to another system as secondary disks and see if you can recover the data.
DO NOT WRITE ANYTHING TO THOSE DISKS until you have at least attempted to make backups of all important data that is still readable on them. First try to mount the disks as read-only and backup everything you can. Only after that you should try to run a filesystem check and see if that helps to recover more data.

Since 2 of your threatened disks form a RAID 1 pair, don't operate on those 2 disks at the same time. If you make a mistake with one of those disks, you might still have a chance to recover your data from the other disk.

MK

[kemo]

thanks