Date: Tue, 28 Jun 2005 15:14:46 -0700 (PDT) From: "Casey Scott" <casey@phantombsd.org> To: "Gary Mu1der" <gmulder@infotechfl.com> Cc: Matt Juszczak <matt@atopia.net>, freebsd-questions@freebsd.org Subject: Re: On recent crashes Message-ID: <14383.199.181.134.212.1119996886.squirrel@mail.phantombsd.org> In-Reply-To: <42C182D9.80801@infotechfl.com> References: <1119950428.22027.1.camel@tarkhil> <C7F95488-8AB6-48D7-85F2-9B400B8F5C31@khera.org> <42C171AA.6090802@atopia.net> <42C173D9.3010408@Rainbow-IT.net> <42C175B6.7060702@atopia.net> <42C1777C.4000307@infotechfl.com> <20050628124701.L70927@neptune.atopia.net> <42C182D9.80801@infotechfl.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> Matt Juszczak wrote: >>> I can't reproduce the crash with a non-SMP GENERIC kernel. >>> >>> Gary >> >> So does this mean my problem is nagios? > > Couldn't say, but I would think not. Nagios should be running as an > unprivileged user so it really shouldn't be crashing FreeBSD. > > However I'm fairly certain I can crash my SMP FreeBSD 5.4 server with > enough ping activity over time, which indicates a bug somewhere in the > network stack (maybe a race condition, given that it requires a SMP > kernel). Nagios with a lot of hosts sends a lot of pings, which would > imply that Nagios could eventually crash the server as well. > > However you said that Nagios process itself died - that would indicate > some other problem w/Nagios. > > Gary > > ps. now testing FreeBSD 4.11 SMP with 130+ simultaneous ping/arp -d's > (50,000 context switches a second!) and it has been rock solid for 30 > minutes. 5.4 would crash within 60 seconds. > _______________________________________________ > freebsd-questions@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-questions > To unsubscribe, send any mail to > "freebsd-questions-unsubscribe@freebsd.org" > I've also experienced 5.4 crashing under network stress with a SMP kernel. In my case, it has with couple simultaneous file transfers via NFS. I have also seen it tip over under high network load from just about anything though. It's also worth noting that I have had interfaces go offline under load. It's not a hardware issue because I have swapped in NICs that I know work in others systems, and still had then go offline. The NICs this occurs to never show any issues under any other OS. The NICs have all been 3Com 905B's. What type of NIC was used with the Nagios issue?
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?14383.199.181.134.212.1119996886.squirrel>