Subject: Re: Web load causes reboot
To: Frank van der Linden <frank@wins.uva.nl>
From: Koos de Haan <sdehaan@caiw.nl>
List: current-users
Date: 08/04/1999 16:05:14
Hi,

When i marked the rnd device out in the kernel and compile a new kernel the server seems to running stable.

I will do some further testing with more and longer load.


Thanks and regards,



Koos de Haan


Op 15:06 +0200 04-08-1999, Frank van der Linden schreef:

> Well, what do you know, I was away for 2 hours and in the meantime someone
> did come up with a backtrace and a reasonable analysis.. (see message
> below if you haven't seen it).
> 
> Let me know if removing the rnd pseudo device from the kernel helps.
> 
> - Frank
> 
> ===============================================================================
> 
> Subject: Re: Web load causes reboot                                             
> Date: Wed, 04 Aug 1999 08:03:09 -0400
> From: Bill Sommerfeld <sommerfeld@orchard.arlington.ma.us>                      
> 
> I just opened a PR on this.
> 
> I *suspect* that the workaround for the "Web load causes reboot" is to
> turn off pseudo-device rnd.
> 
> So, what's going on here is that rnd.c assumes that splsoftclock()  is
> like other splfoo() calls in that it masks off interrupts.
> 
> This is not, in fact, the case.  It *enables* interrupts, and thus
> *un-protects* a chunk of code from interrupts.. in this case, against
> a recursive softnet() interrupt.
> 
>                                         - Bill
> ===============================================================================


-
--

Kabelfoon BV
Industriestraat 30, Postbus 45, 2670 AA  NAALDWIJK
Tel. 0174-640269, fax 0174-623860
e-mail info@kabelfoon.nl voor informatie, kabelfoon@kabelfoon.nl voor abonnementszaken, help@kabelfoon.nl voor de helpdesk
Zie http://www.kabelfoon.nl/pgp/ voor Kabelfoon public keys