Hi Everybody!

I got the following message on my Redhat linux box:

I would appreciate if somebody could let me know the following:
1) How can this happen?
2) Is there a way to fix it?

################################
Journal_bmap_Rsmp_6199acb1: Journal block not found at offset 7180 on sd(8,2)
Aborting journal on device sd(8,2)
ext3_abort called
EXT3-fs abort (device sd(8,2)):ext3_joural_start: Detected aborted journal
Remounting file system read-only
EXT3-fs error(device sd(8,2)) in start_transaction: Journal has aborted
###############################

I tried boot in rescue mode and tried, fsck, badblock checks(no badblocks found). But when I run linux i get the message above everytime.

Vincent

If you use an usb memory stick and an entry in the /etc/fstab or /etc/mtab to auto mount it you can get this message if the usb stick is not in working order.

use dmesg to see what is wrong

################################
Journal_bmap_Rsmp_6199acb1: Journal block not found at offset 7180 on sd(8,2)
Aborting journal on device sd(8,2)
ext3_abort called
EXT3-fs abort (device sd(8,2)):ext3_joural_start: Detected aborted journal
Remounting file system read-only
EXT3-fs error(device sd(8,2)) in start_transaction: Journal has aborted
###############################

I tried dmeg and it outputs errors about EXT3-fs, like as above.

As you can see in the error above, which is the first error you get to see on the screen when you boot up the server. The filesystem is automaticly set to "read-only". A server with a read-only files system cant do much. I am not able to write anywhere on the filesystem to any file.

And mounting an USB device, is not really what i need. I would like to get the files system mount again as "read-write" and get the server running again. But somehow there is a problem with EXT3-fs or the Harddisk, I think...Hope somebody knows about this problem and can tell me what it is...

A sd can also be a scsi block device (disk).
During a boot of a server you can access a scsi controller menu.
Look if there is an option to check the disk

start in single user

init s

and manually check disk

fsck

Thank you for your reply,

I tried boot in rescue mode and tried fsck. Of course there came some errors about blocks and inodes. On all questions I did "Y" for fixing it. But after the reboot same errors start to appear and the filesystem is switch to read-only.

Though the name of the drive is "sd" the harddisk is not a scsi drive. Its a 10.000 rpm HDD ide.
(and with a raid controller) Because there two swapeble disks of those. But I used only one of them. Which is now broken probably...as I cant really fix it.

I reinstalled the system on the other disk. But am still wondering what happened on the first disk...and if it is actually fixable or better to throw away the disk...

Vincent

What kind of raid level do you use level 0 or 1.
Every raid controller comes with software to check the disk

Hi Arno,

I did not use the raid. I used only one disk.
But because we had to get the server online again, I used the second disk to reinstall the system.

So I cant check the controller menu, for any checking options, because that means I have to shut down the server and we will be offline. I cant do that right now. (but I will take a look into the manual and prepair how to get into the menu and check the disk, so that we can be for a short time offline, just to check it to be sure, but I have to be well prepaired so that it takes little offline time).

Thats why I was hoping to get some answers here...so that we dont have to be offline again for researching the old disk.

Thanks,
Vincent