Help with PANIC error message

[fathead]

Panib : HTFS :BAd directory ino 41489 (offset 728111) on HTFS dev hd (1/43)

trying to dump 261999 pages to dumpdev hd (1/41) 13275 per `.`

261999 pages dumped
NOTICE: Fluching perc adapter

I got the error this morning while...any ideas...

Dell Poweredge 2600
2.8 Zeon
1 GB Ram
3 18GB drives (RAID 5)
Perc 4/di scsi adapter
SCO 5.0.7 with maintenance pack 1 installed

 

[PHV]

The 3rd filesystem (division) of your primary disk is corrupted.
To retrieve their name:

fs=`echo "q\ne" | divvy | awk '$8==3{print $2}'`

or:

fs=`ls -l /dev | awk '/^b.* 1, 43 /{print $NF}'`

In Single User (maintenance) mode do:

fsck -ofull /dev/$fs

You will probably have to take a look in the

 lost+found

directory of this fs.
Anyway, I hope you have reliable backup.


Hope This Help
PH.

 

[fathead]

OK...This is a new install so no data on server yet...

I decided to reinstall SCO...no problems with install or installing any updates, service packs, or software.

I continue to have kernel panics...I don't have the exact error messaage, but it is similar to one above (I will post exact error message tomorrow...was working on remotley).

It seems to panic when I am accessing /u filesystem...Friday, I removed /u then re-added...seemed to be OK..started coping data from the existing server to /u.

Saturday, I logged in remotely and started a program on /u..this is when it had the panic.

Any suggestions???

 

[fredericofonseca]

Option 1- Change Raid controller
Option 2- Change HD's

One of them is most probably giving you the error.



Regards

Frederico Fonseca
SysSoft Integrated Ltd

http://www.syssoft-int.com

 

[PHV]

Have you tried this ?
In Single User (maintenance) mode do:
fsck -ofull /dev/u
You will probably have to take a look in the lost+found directory of this fs, after mounting it.


Hope This Help
PH.

 

[fathead]

I did fsck -ofull /dev/u and it corrected several problems.

I tried this before I reloaded SCO and it found problems then too...bad hard drive?????

 

[PHV]

Not necessarly bad disk, but surely bad FS, regards the PANIC.

 

[melissamay1]

Hi there:

Too bad for the Panics, but good that it's a new install...

I've had similar problems on new installs. For me it's usually been some bad drive-related thing, like a bad cable or bad drive.

I'd test the drive subsystem (drives and controller) as much as possible. Swap drives out, if you have extras. Run manufacturer's diagnostics, even if you need to install another non-SCSI drive into the system temporarily (with another OS) to run the tests. Eliminate everything but the drives - remove tape drives and CD drives from the SCSI bus if they're there.

I once had to abandon 3 SCSI drives and a RAID controller because nothing would get them to work on SCO. I ended up just using them for another project...

Good luck!

 

[PHV]

Another tip:
man badtrk

Hope This Help
PH.

 

[fredericofonseca]

This is a SCSI setup, so badtrk should have no effect.

as I said, and melissamay1 reiterated, try and change Controller and/or HD's. (Controller first)
try the hd's with no raid and raid 1.

Check cables and terminator (ACTIVE TERMINATORS ONLY!!!)

If this is a Ultra320 setup this may also be the issue.




Regards

Frederico Fonseca
SysSoft Integrated Ltd

http://www.syssoft-int.com

 

[fathead]

Hers's the error message...

Panic: k-trap - kernel mode trap type 0x0000000E trying to dump 261999 pages to dumpdev hd 1/41 3275 pages per '.'

??????

 

[PHV]

From man trap :
0x0E PGFLT fault page fault; page could not be brought into
memory due to bad memory, or a bug in a third party device driver

Hope This Help
PH.

 

[apeasecpc]

Filesystems don't normally have errors right after you set them up on a fresh install.

Your problem is a bad controller, bad hard drive, or bad controller driver. Are you sure you are installing the correct driver for your version of SCO? Drivers for an old version will not necessarily work correctly, and I seem to recall some previous posts of problems with Perc 4.

 

[fathead]

Well, I finally removed the RAID config, and setup the three drives as standard drives. I installed the scsi driver from the Lsi logic website and, so far, have not had a problem. It would have been nice to have the RAID working, but the speed without it will work. I really appreciate all the input!!! I will keep you posted as to the status in the next couple of days/weeks...