Diffrence between fsck on HP11.0 & HP11.11

[Juan Gonzalez_2]

What I am tring to do is mount a mirror copy of my production database on a backup system but the fsck goes into a loop looks like and it never can fix the file system on a system that is running HPUX11.11 but if I try to mount it on a HPUX 11.00 server it fixes the filesystem and mounts the volume group. See attached log file. Can any ong help? I need to move this mirror into the new HPUX 11.11 server because the old server running HPUX11.0 is going of lease.

[LucianoCarvalho · Accepted solution]

hi juan!

What is the version of jfs on 11.11 machine and 11.00 machine? They could be running diferent versions.
HP-UX 11.11 comes with jfs 3.3 as default and 11.00 don't.
If you run "fstyp -v /dev/vgxx/rlvolx" on hp-ux 11.11 you will see version 4 in the output, that is the layout version availabe with jfs 3.3.
On hp-ux 11.00 you will see version 3 in the output.

You can install jfs 3.3 on 11.00. See the link below.
http://www.software.hp.com/cgi-bin/swdepot_parser.cgi/cgi/displayProductInfo.pl?productNumber=B8861E

Regards

[Juan Gonzalez_2]

Luciano,
The version of online JFS is the same on bothe servers because I had to install online JFS 3.3 on the HPUX 11.0 server for some other requirment. See output below:

vxfs
version: 4
f_bsize: 8192
f_frsize: 8192
f_blocks: 11520000
f_bfree: 3000908
f_bavail: 2977465
f_files: 807040
f_ffree: 750224
f_favail: 750224
f_fsid: 1077018625
f_basetype: vxfs
f_namemax: 254
f_magic: a501fcf5
f_featurebits: 0
f_flag: 16
f_fsindex: 6
f_size: 11520000

[Juan Gonzalez_2]

Also the version of fsck on HPUX 11.0 is $Revision: 92453-07 linker linker crt0.o B.10.60 970821 $

On HPUX 11.11 the version of fsck is $Revision: 92453-07 linker linker crt0.o B.11.16 000601 $

[John Poff]

Hi,

I've done lots of this stuff before in scripts and I've never seen any fsck errors like that. Can you post your script that does all this and let us take a look at it? It might give us some clues about what is happening.

JP

[LucianoCarvalho]

It's really strange Juan.

I don't know if you already, but if not, I think it's worthwhile
to install the last patches for fsck on both machines.




Patch Name: PHCO_24036

Patch Description: s700_800 11.00 fsck_vxfs(1M) JFS 3.3 cumulative patch

Symptoms:
PHCO_24036:
( SR: 8606131436 CR: JAGad00586 )
fsck_vxfs(1M) goes into infinite loop
as follows:
# fsck -F vxfs -o /dev/vg01/rlvol2
replay in progress
file system is not clean, full fsck required
pass0 - checking structural files
vxfs fsck: structural inode 97 (Primary Ilist 1) failed
validationclear? (ynq)y
pass1 - checking inode sanity and blocks
pass2 - checking directory linkage
pass3 - checking reference
countsrebuild structural files? (ynq)y
pass0 - checking structural files
vxfs fsck: structural inode 97 (Primary Ilist 1) failed
validation clear? (ynq)y
Pass2 ... .





Patch Name: PHCO_24813

Patch Description: s700_800 11.11 fsck_vxfs(1M) directory corruption


Patch Name: PHCO_26701

Patch Description: s700_800 11.11 fsck(1M) sync switch patch

[Decio Miname]

Hope I am wrong and telling you something you already knew, but...

When you "broke" the mirror from the production VG, thus releasing the mirror copy to be mounted anywhere else, were the production databases down?

If they were up and changing data to disk, you'll probably never have the filesystems consistent enough to be mounted anywhere, that's why fsck would be facing so many errors.