On Sep 24, 2:18 am, Alberto Rodriguez wrote:
> Thirty two days up and running!
>
> No more hangs until now! :-)
>
> --
> Alberto Rodriguez Rodriguez

I am very happy as well. I found an extra download on HP's site. HP
Networking Pack, I downloaded the pack looked into my NIC driver's
folders and found a new driver version.

The driver version included with EFS was 2.8.7 the driver version in
the pack was 3.4.0.

The new driver has been installed and I have an uptime of 12 days
now. I haven't been online for 12 solid days since the first month of
the machine being online.

I'll continue to update after I have 1 month uptime.

-David

I have a customer with SCO OpenServr 6/mp2 also with random hangs.
It is an HP Proliant ML-350G5 w/3GB & SAS Raid - i've setup similar
a few times, yet this one in particular randomly hangs as well.

the only thing different about this install others i've done with identical
hardware/software is their reliance on terminals (Digiboard C/X, maybe 30 to
40 terminals & serial printers) and mscreen. because of this i contacted
digi tech suport, who insist its a sco problem.

users log out & never get back to login, who shows them still logged in,
eventually all are hung. could take anywhere from 1 day to 3 weeks after
are reboot before this happens.

no error messages to console, nothing in syslog/messages, no indication
of running out of any resources, etc. built-in hp diagnostics run & don't
find any errors.

I tried the KHZ=100 thing and it did not help in my case at all.

ideas?

-joe

> Mike wrote:
> >
> > Check the file /stand/boot. Make sure a line "KHZ=100" is there, or
> > add it and reboot.
> >
> > Mike
>
>
> Following the link supplied by John, there is another link:
> (http://www.sco.com/ta/126735) relative to UnixWare 7.1.4 in wich
> it's explained the behaviour of the parameter KHZ (also valid for
> OSR6).
>
> This parameter is modified from KHZ=100 to KHZ=1000 in OpenServer
> 6.0.0 when patch OSS706B or OSS706C are loaded (not OSS706A).
>
> It is known that this change "... may be related to some system hangs
> or reboots reported to SCO, as described above. These issues are under
> investigation."
>
> Last ptf9052g for UW 7.1.4 resets this parameter to its original value
> (KHZ=100)
>
> So I agree with Mike. This may be the origin of the trouble. I've
> added KHZ=100 to /etc/default/boot, and the server will be rebooted
> this midnight. I must wait for resluts in the next days. I'll post the
> results.
>
> To all that speak about memory problems, I forgot to say that memory
> in this server is ECC memory from HP, 6 DIMMs, 512 MB each, DDR2.
> Memory, motherboard, RAID controllers, etc. have been tested with test
> programs supplied by HP without errors (not tested 96 hours as Boyd
> proposed...)
>
> Thank you all again. I'll keep you informed.
>
> Best regards.
> --
> Alberto Rodriguez Rodriguez
>
>
>
>
>
>
--- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ---
-Joe Chasan- Magnatech Business Systems, Inc.
joe@magnatechonline.com Hicksville, NY - USA
http://www.MagnatechOnline.com Tel.(516) 931-4444/Fax.(516) 931-1264

Hi all,

Two months after changing parameter KHZ=100 in /etc/default/boot, the
system has not crashed at all. It's up and running without problems!!!

I want to give a big "Thank You" to all of you who shared your
knowledges and ideas. Your help has been very valuable.

Joe Chasan wrote:

> I tried the KHZ=100 thing and it did not help in my case at all.
> ideas?

Joe,

I had the same problem (random hangs) with another two servers with a
very different hardware. Both of them had one hang after chanching
KHZ=100, but I made a full fsck of filesystems and from that moment
on, they work flawlessly. No hangs.

As single user:

# /etc/fsck -y -ofull /dev/root

# /etc/fsck -y -ofull /dev/u ...or whichever other filesystems

Try to assign a single (not shared) IRQ for the Digiboard card in
server's BIOS.

Hope this helps. Best regards.
--
Alberto Rodriguez Rodriguez