Too much work at interrupt

[Daniel Veillard]

 Like Michael Frank (mfrank@allot.com), I got this message while load-testing
one of our new Web servers. But unlike the previous example, the machine
stayed up, I just had to do a /etc/rc.d/init.d/network stop and start
to get the machine back online. It's an HP LH Pro, dual PPro200, with 2
EEPro/100, only the first one is used yet and plugged directly to a 100MBps
switch (Bay Network) 100MBits detected and full-duplex.

 I'm using a kernel 2.0.35 patched to version 1.03, the same I started using
on our VAResearch servers (Hardware raid), but those servers never showed this
message (while they show the "SMP simultaneous entry of an interrupt handler"
occasionnaly but apparently without damage on 1.03).

 Any chance that both are variations of the same 2.0.xx SMP kernel bug ?
Shoudl I upgrade to 1.05 ? Should I try to increase the 
  static int max_interrupt_work = 20;
increase as suggested in a past mail ? 
Or is there a way I can help debug this thing before the machines goes
into production ?
 Our servers are all using the EEPro100 and that's the first time I see
this, is it just bad luck of something related to some hardware setup ?
I had pushed the machine under a double 100MBps ping flood without seeing
the error which occured only when starting pushing the HTTP and AFS setup.

  thanks for any insight,

Daniel

-- 
Daniel.Veillard@w3.org | W3C  MIT/LCS  NE43-344  | Today's Bookmarks :
Tel : +1 617 253 5884  | 545 Technology Square   | Linux, WWW, rpm2html,
Fax : +1 617 258 5999  | Cambridge, MA 02139 USA | badminton, Kaffe,
http://www.w3.org/People/W3Cpeople.html#Veillard | HTTP-NG and Amaya.