SCIENTIFIC-LINUX-USERS Archives

September 2006

SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Michael Mansour <[log in to unmask]>
Reply To:
Michael Mansour <[log in to unmask]>
Date:
Sun, 24 Sep 2006 12:14:06 +1000
Content-Type:
text/plain
Parts/Attachments:
text/plain (58 lines)
Hi Connie,

> > Hi Connie,
> > 
> > > > Hi,
> > > > 
> > > > I've recently rebuilt some of my DL360's from earlier releases to SL4.3
> > > > 
> > > > With the HP Proliant DL360's (generation 1) single CPU units, they
refuse to
> > > > boot the latest 2.6.9-42.0.2.EL kernel, but boot fine from the earlier
> > > > 2.6.9-34.EL kernel.
> > > > 
> > > > There's no kernel panic from the boot screen, just hangs on boot and the
> > > > machine has to be power cycled.
> > > > 
> > > > On exactly the same model servers but with an extra CPU, booting off the
> > > > latest SMP kernel boots fine.
> > > > 
> > > > Any ideas how I can troubel-shoot this one?
> > > > 
> > > > I've yet to disbale the "quiet" boot mode to see exactly where it
hangs. I've
> > > > also yet to try to boot an SMP kernel on a UP server to see if it would
> > > > actually work.
> > > 
> > > If hyperthreading is turned on try turning it off.
> > 
> > There's no hyperthreading on these machines. 1Ghz PIII cpu's.
> > 
> > > Also could try
> > > 
> > >    linux noapci
> 
> Sorry that should be
> 
>        linux acpi=off
> 
> or another one to try
> 
>        linux noapic
> 
> I get those backwards all the time.

Adding the noapic allows it to boot the kernel. But isn't this important to be
able to monitor the internals of the machine?

I use the hp proliant support pack on these proliant servers, which has a
hpsmh health monitor which monitors the hardware and alerts on problems. If
this becomes ineffective with the disabling of the acpi facilities, then I
would rather move back to the earlier kernel which works.

Obviously that isn't a long term solution, but I'd also raise a case in RH's
bugzilla (referencing RHEL43 - I have over 60 RHEL subscriptions with them
anyway) to see if they can resolve it.

Michael.

ATOM RSS1 RSS2