Date: Tue, 8 Apr 2003 06:50:22 -0700 From: Luigi Rizzo <rizzo@icir.org> To: Steve Shorter <steve@nomad.tor.lets.net> Cc: freebsd-net@freebsd.org Subject: Re: poll stalled? Message-ID: <20030408065022.A4568@xorpc.icir.org> In-Reply-To: <20030408133419.GA17964@nomad.lets.net>; from steve@nomad.tor.lets.net on Tue, Apr 08, 2003 at 09:34:19AM -0400 References: <20030408133419.GA17964@nomad.lets.net>
next in thread | previous in thread | raw e-mail | index | archive | help
inaccessibility is (possibly) coming from a bug in setting the watchdog in the fxp (and other) drivers, long since fixed I believe. On top of this, the '...stalled' messages come out when some interrupt handlers take longer than one timer tick, something that happens frequently e.g. when there is a change in the PHY of the medium, and also for other reasons. cheers luigi On Tue, Apr 08, 2003 at 09:34:19AM -0400, Steve Shorter wrote: > Howdy! > > I have a server, FreeBSD 4.6.2, with 3 fxp network interfaces, > polling enabled, kern.polling.user_frac is 50, and kernel > compiled with Hz = 1000. > > It became inaccessable over the network and when I logged in > over the console the following message(s) were written to it... > > : > poll [14322] stalled in phase 2 > poll [14323] stalled in phase 2 > : > > etc.. > > What causes this? Is it a bug or an indication for tuning? > The only tunables for polling that I know of is Hz and the user_frac sysclt > > > Is there a fix for this? > > thanx - steve > > _______________________________________________ > freebsd-net@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-net > To unsubscribe, send any mail to "freebsd-net-unsubscribe@freebsd.org"
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030408065022.A4568>