Update: PowerEdge6400 PCI Interrupt Bug

Curtis Rubel crubel at compro.net
Wed Mar 27 16:37:00 CST 2002


Hello all once again,

        After reporting the situation to Dell, the only
recommendation was to upgrade to the Enterprise Server
version of Redhat.  They seem to think that they kernel
that ships with the regular version will not properly support
more that 2 cpus...
        
        I seem to recall that the Linux kernel will support
up to 128 cpus out of the box, but I am not 100% positive
on the number.

        At any rate, we currently have all 4 cpus reinstalled, running
the same 2.4.18 kernel we had previously, our PCI cards are
installed in slots 1,2,5 and the system is running ok.  We will leave
it cycling overnight to verify things.  

        So either we have a bad PCI slot on this motherboard or
there is some other issue.  I am hoping Dell will send in another
Motherboard so we can verify this one way or another before
shipping this to our customer.

If anyone has seen or heard of any problems like this I would 
appreciate any information that you could provide.

Regards,






-----------------------------------------------------------------------------------
We have recently discovered what appears to be
a PCI Interrupt bug in the PowerEdge 6400 Server system.
When we have 3 PCI boards installed into slots 2, 5, and 6,
the system will lock up during periods of PCI interrupt activity.
It appears if we removed the board from slot 6, the system 
runs ok. Reinstall that same board in a different slot and it
runs ok..any board installed in slot 6 and the system hangs.

We removed 2 cpus and installed the required terminator
boards and retested. All boards installed as above and the system
is running fine, removed the 2 cpus and installed the other 2 cpus
just to make sure we do not have a bad cpu, and the system still
runs ok.

We have already had the motherboard replaced once, but
I cannot be sure that this is not still a motherboard problem, since
we were not down to this level before the original motherboard was
replaced.

We are running the vanilla Linux kernel 2.4.18 at this time, system
has 1GB of memory, system bios is at rev A10.

I just wanted to get this posted in case there is someone else
with a similar problem. If/When we get this resolved I will post the
fix for all.



Curtis Rubel                                   Compro Computer Services, Inc.
Product Development Engineer   105 East Drive
Office: (321)727-2211                  Melbourne, FL 32904
Cell: (321)258-9348                      Email: curtis at compro.net




More information about the Linux-PowerEdge mailing list