Re: sysctl spinning (was: Re: ps Causes Hard Hang)

From: Bernd Walter <ticso_at_cicely12.cicely.de>
Date: Thu, 4 Mar 2004 06:03:11 +0100
On Wed, Mar 03, 2004 at 01:38:27PM -0500, Robert Watson wrote:
> 
> On Wed, 3 Mar 2004, Cy Schubert wrote:
> 
> > I'm running 5 -CURRENT systems. My firewall system, using IPF, hard
> > hangs every time ps is entered -- totally unresponsive, requiring either
> > a power cycle or reset switch to bring it back to life. 
> > 
> > Before I start digging into this seriously I'd like to possibly get info
> > from anyone who may have experienced this before. 
> 
> Alan Cox and I have both experienced this -- it's actually only a hard
> hang if you're trying to use the syscons break to debugger, serial break
> to debugger can get into DDB fine.  It looks like the sysctl code is
> spinning in kernel, possibly due to looping waiting for a response other
> than EAGAIN.  I'm wonder if it was the recent limits on locked memory
> changes in sysctl, although at first we thought it might be the sleepq
> changes (seems less likely now).  Because sysctl holds Giant, the other
> CPUs are locked out of Giant-protected bits of the kernel (many of them),
> including Syscons.

Me too.
My system is also a router, but running ipfw.
At least ps, top and w show this problem.
System is syscons so I wasn't able to get into DDB - unfortunately all
sio(4) serials are half-duplex RS485 so I can't switch to serial console
without swapping cards.
All commands work fine for a while if I reduce the number of services
startet at boot time and it never happened in single user mode.
I couldn't make out a single service to trigger this problem.

-- 
B.Walter                   BWCT                http://www.bwct.de
ticso_at_bwct.de                                  info_at_bwct.de
Received on Wed Mar 03 2004 - 20:03:34 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:46 UTC