I am running several (15) PCengines WRAP boards with FreeBSD and M0n0wall 
and having the same freezes.  I thought it was power; put it on UPS and 
froze.  I thought it was heat; added fans to my wifi box enclosures, still 
freezes.  I am not having trouble where my user levels are below 10, but 
where I can see 20+ users I am having the freezes.  I also see massive 
amounts of P2P traffic before freezes, but even with limiting the P2P 
traffic to 5% on the Traffic shaper it still freezes (though a bit less 
often).  I am seeing one error that I can't explain or make go away and that 
is numerous "stray IRQ 7", but I can't explain that one either.

I am about to change to Microtik Router boards using the RouterOS.  They 
have a different CPU chip on their boards and I am hoping that will solve 
the problem.  Talking to them they say they can handle 100 simultaneous 
users as well.  I am also today going to test Ikarus OS on one of them.  I 
have a theory it is the small CPU on my WRAP boards that can't handle the 
excessive traffic and seems to lock up the boards.

I have not tested M0n0wall on PCs because that is not where I need it to 
run.  I am running wireless hot spots so I need the small form factor boards 
with wireless capabilities.  I have to admit that on my low traffic sites 
M0n0wall has never had to be rebooted; but on my heavy use sites keeping 
them running is a constant up hill battle.

I have very little hair left as well!

Brad Shows
Americable Int'l
System Administrator

----- Original Message ----- 
From: "Aaron Cherman" <aaronc at morad dot ab dot ca>
To: <m0n0wall at lists dot m0n0 dot ch>
Sent: Sunday, August 27, 2006 7:09 AM
Subject: Re: [m0n0wall] m0n0s hanging : any progress ?


>> My m0n0 hang this morning at 5am, nothing in my remote syslog. I still
>> don't think that is setup correctly as I don't get much more than packet
>> rule information. Regardless. Hard locked, snmp stopped responding,
>> reboot and it came right back. I am getting ready to switch over to
>> pfsense and see if I have any better luck. If not, I am going to attempt
>> to set up a cluster with two pfsense boxes with one for failover. We'll
>> see how it goes.
>
> I have made a number of changes that all affect P2P traffic - using my 
> NetEnforcer I have forced all P2P traffic (that can be classified) into a 
> single virtual channel that I can work with.  I have reduced the max 
> number of live connection to 80 (only for P2P) and reduced the max 
> connections initiated per second to 15.  I have also limited the P2P 
> traffic to 200 kbps up and 200 down.  I still have the nightly reboots, 
> and this morning at 6:15 this unit froze up.  I have absolutely no hair 
> left after pulling it all out.  I have also just submitted a purchase 
> order this afternoon for a Cisco router.  m0n0wall has beaten me.  I give 
> up...    :-(   I am almost convinced that P2P traffic has nothing to do 
> with the freezes, but where do I go from here?.....
>
>
> Aaron
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: m0n0wall dash unsubscribe at lists dot m0n0 dot ch
> For additional commands, e-mail: m0n0wall dash help at lists dot m0n0 dot ch
>
>
>