heavy load crashes network services

gil gil@truepath.com
Thu May 27 15:16:23 1999


Hi,

I'm running Red Hat Linux 5.2 (linux 2.0.36) on an IBM 704 quad
processor box (only two ppro processors are installed). I have tried
both the Ethernet Express Pro and the 3COM 3c905B (Vortex) cards. Both
cards are crashing the system or giving error messages during heavy load
on our main web server.

The eepro card running driver eepro100.c 0.99B resulted in a system
crash with the following error:

May  1 09:05:45 reliant kernel: eth0: Transmit timed out: status 7048 command 0000.
May  1 09:05:45 reliant kernel: eth0: Trying to restart the transmitter...

I upgraded the driver to v1.05 and under heavy load the system stays up
with all processes running, but the card no longer functions. i.e.,
cannot ping any addresses. The same error is displayed as above.

Then when, I tried the 3COM 3c905B card with the 3c59x.c  0.99H-WOL
driver, I received this error which was repeated every few seconds:

May 26 19:43:55 reliant kernel: eth0: transmit timed out, tx_status 00 status e601.
May 26 19:43:55 reliant kernel: eth0: Interrupt posted but not delivered -- IRQ blocked by another device?

The web server still ran, but it was just crawling along very slowly.

After much research, it seems that having SMP enabled may be part of the
problem. I will disable it and see what happens.

What else should I try? Is Linux 2.2 able to cooperate better with the
the eepro100 and 3c905 cards under heavy load? Any help/hints would be
appreciated.

in His grip,
Gil
http://www.truepath.com
your Christ-centered web host