MP S2460 Problem

Maurice Hilarius maurice at harddata.com
Thu Feb 27 00:09:41 PST 2003


With regards to your message:
>From: "Robert G. Brown" <rgb at phy.duke.edu>
>To: Ken Chase <math at velocet.ca>
>Cc: beowulf at beowulf.org
>Subject: Re: MP S2460 Problem
>
> > Oh do you mean PCI riser cards to let 2U servers have full height
> > PCI cards in them? Bad karma on those - we've used them in a few
> > machines around here and we've had problems with almost every board
> > we've tried them on. PCI doesnt like to have extra goop in the way of
> > its cards.
>
>Yes, that kind of riser.  And we finally did find risers that worked
>adequately.  The PCI bus on the 2460 has (I would guess) some critical
>timing issues with cards.  Some cards that work fine in any OTC box just
>won't work or will instantly crash the box (preboot or midboot).  They
>were incredibly picky about what slots would work with even cards that
>DID work.

The correct answer to a great degree is to notice that Tyan makes a 2U 
riser card for the S2460,62, 66, 68, 69, called the M2039

With this card in use we have found Tyan to be VERY diligent about support 
for issues on it.
Case in point:
A customer of ours last year wanted to have 2466N-4M with a 4 port ethernet 
card ( DLink using 4 "tulip" chips", video, and we ran into some issues.
Problem is the riser is essentially a PCI bridge.
The DLink 4xNIC has a PCI bridge on it.
IRQ conflicts occurred.

The Tyan tech support spent a good amount of time to get the parts we were 
using, and solved the problems.

Case closed, over 80 of those machines now filter traffic for Internap 
around the USA and the world.

>We worked our way through various BIOS flashes, and have them running as
>stably as possible (for us) now.  They don't generally crash under no
>load, they don't generally crash under MY application, but certain
>applications that run like a charm on P4's or 2466's crash the
>essentially IDENTICALLY configured 2460's as described.

And with current BIOS, proper riser, and appropriate settings we see 
nothing comparable to that on 2466 and later.
Frankly, the earlier AMD MP chipset found in the 2460 and 2462 is not as 
stable as the newer MPX, but in general it is not as bad as you imply. IMHO.

>Tyan refused to own the riser problem OR the card problem.  It was just
>plain the fault of the card manufacturer, not theirs.  Damn those
>incompetent folks at 3com, anyway.  Strange how they opted to use their
>onboard chipset for the same device on the 2466.
"The riser card problem" Meaning "somebody else's riser card problem", I 
presume?

> > WOW. Thats amazing. We can run them without keyboards fine. Occasionally
> > the BIOS is reset by as power flux or something like that, and yes
> > you have to get out the video card and stuff, but thats a minor deal.

We ran into this, and asked for a customized BIOS version with defaults to 
serial console, ignore VGA and keyboard errors, and they have complied.

Perhaps they are more inclined to provide this support to their vendors 
than end users, I do not know..

Don't take me wrong, they are not perfect.
However, when we run into issues they have always been good about solving 
them for us, anyway. Maybe it's my friendly face..



With our best regards,

Maurice W. Hilarius       Telephone: 01-780-456-9771
Hard Data Ltd.               FAX:       01-780-456-9772
11060 - 166 Avenue        mailto:maurice at harddata.com
Edmonton, AB, Canada      http://www.harddata.com/
    T5X 1Y3




More information about the Beowulf mailing list