Very Computer

I apologize if this is in the wrong group...but I have a small
problem. Some filesystems got corrupted on one of our R6000
and I ran the diagnostics to correct it/them which it seems it
did, but there is still one filesystem (lv00) that will not mount
when the machine is rebooted.  I've tried using fsck to correct
it but it doesn't do any good...it says something about duplicate
inodes and fsck stops just after phase 1b.  If I try to run
fsck -y -i# /dev/lv00 it references the inode I used in the command
line and lists a file ( /log/tps.stderr ) that I guess is linked to
it.  The problem is I can't see /log and can't even find the file
tps.stderr anywhere on the system.  Is there some command or variation
of fsck that will eliminate duplicate inodes.

Also, when booting, the terminal reports that the lv00 filesystem
superblock is dirty and that I need to run a full fsck on it to fix
it. That's when I get the duplicate inode messages in phases 1 and 1b.

I would really appreciate some help and am trying to do everything
possible to gradually learn Unix and AIX.

Version is 4.1.2 (I think)
hdisk0 and hdisk1 are each 400MB internal hard drives
hdisk2 is external 16bit 4.5GB

Thanks,
John

do you get any LED's on startup ?

Surprised you didn't mention the "fsck" bug in AIX 4.2.1.

Anyway, "fsck" should always go through all phases unless you abort
or run it with limited options.

There is a bug in AIX 4.2.1 for "fsck" that will cause it to core dump.
You will see them in your error report.

I don't have the APAR number, but you can look this up on IBM's
web site or call IBM and order it.

 >> I apologize if this is in the wrong group...but I have a small
 >> problem. Some filesystems got corrupted on one of our R6000
 >> and I ran the diagnostics to correct it/them which it seems it
 >> did, but there is still one filesystem (lv00) that will not mount
 >> when the machine is rebooted.  I've tried using fsck to correct
 >> it but it doesn't do any good...it says something about duplicate
 >> inodes and fsck stops just after phase 1b.  If I try to run
 >> fsck -y -i# /dev/lv00 it references the inode I used in the command
 >> line and lists a file ( /log/tps.stderr ) that I guess is linked to
 >> it.  The problem is I can't see /log and can't even find the file
 >> tps.stderr anywhere on the system.  Is there some command or variation
 >> of fsck that will eliminate duplicate inodes.
 >>
 >> Also, when booting, the terminal reports that the lv00 filesystem
 >> superblock is dirty and that I need to run a full fsck on it to fix
 >> it. That's when I get the duplicate inode messages in phases 1 and 1b.
 >>
 >> I would really appreciate some help and am trying to do everything
 >> possible to gradually learn Unix and AIX.
 >>
 >> Version is 4.1.2 (I think)
 >> hdisk0 and hdisk1 are each 400MB internal hard drives
 >> hdisk2 is external 16bit 4.5GB
 >
 >Could you post the exact errors you are getting?  You might be in a
situation
 >where the file system can't be fixed, but it's worth a chance seeing the
exact
 >errors...