Hi all,

I'm new to LINUX, though I have been an HP-UX admin for 12 years. As it stands, I've also been a LINUX admin for 8 weeks now. :-) This is a weird problem, so please bear with me.

We're using Red Hat 6.2 (yes, I know it's wayy old), and exporting a volume using NFS to our HP systems. Over the last few days, we've had an issue crop up during our nitely database exports from the HP's to the linux box. When cron tries to copy a file over to linux, the HP system generates errors that the NFS on linux is full:

Aug 26 01:09:25 redbaron vmunix: /tmp_mnt/backup: file system full

However, a quick look at the linux box shows the filesystem to be at roughly 70% or less:

LINUX$ df /storage/redbaron
Filesystem 1k-blocks Used Available Use% Mounted on
/dev/hdb6 8064272 4878396 2776224 64% /storage/redbaron

So I checked the /var/log/messages file, and found the following lines, but don't have a clue what they mean:

Aug 26 02:22:26 taz kernel: EXT2-fs error (device ide0(3,70)): ext2_free_blocks:
Freeing blocks not in datazone - block = 942832688, count = 1
Aug 26 02:22:26 taz kernel: EXT2-fs error (device ide0(3,70)): ext2_free_blocks:
Freeing blocks not in datazone - block = 892481076, count = 1
Aug 26 02:23:59 taz kernel: EXT2-fs warning (device ide0(3,70)): ext2_free_block
s: bit already cleared for block 1712076
Aug 26 02:23:59 taz kernel: EXT2-fs warning (device ide0(3,70)): ext2_free_block
s: bit already cleared for block 1712076
Aug 26 02:30:00 taz CROND[10971]: (root) CMD ( /sbin/rmmod -as)
Aug 26 02:39:52 taz kernel: EXT2-fs error (device ide0(3,70)): ext2_new_block: Free blocks count corrupted for block group 5
Aug 26 02:39:52 taz last message repeated 2 times
Aug 26 02:40:00 taz CROND[10973]: (root) CMD ( /sbin/rmmod -as)
Aug 26 02:40:02 taz kernel: EXT2-fs error (device ide0(3,70)): ext2_new_block: Free blocks count corrupted for block group 4
Aug 26 02:41:30 taz kernel: EXT2-fs error (device ide0(3,70)): ext2_new_block: Free blocks count corrupted for block group 5
Aug 26 02:47:19 taz kernel: EXT2-fs error (device ide0(3,70)): ext2_new_block: Free blocks count corrupted for block group 4


Am I looking at a disk failure? Running out of inodes? Bad superblock? Any and all ideas/suggestions welcome.

Thanks,

-Rich

Take a look here
and check your kernel to see the requirement to run NFS.
Some versions of are Red Hat 6.2 (2.2.14-5).
At least 2.2.18 is needed to run NFS.
check with
ls -l /usr/src/linux

Thanks WHK,

Guess I should've mentioned this in my original post, but my mind was elsewhere. Growing old, I guess..

Anyway, this NFS moint point setup has been working for over 2 years. We also have 3 other linux boxes (all RH 6.2) that serve as NFS points for the HP systems. Everything has worked fine on all 4 boxes until this week, and even then only on this one box.

We've moved the export of our database from this linux box to another one temporarily, until I figure out what the problem is. So, given that it's worked fine for 2+ years, and seeing the errors that I'm getting, any other ideas?

Many thanks,

-Rich