Date: Tue, 27 Mar 2001 17:27:13 -0500 (EST) From: Matt Piechota <piechota@argolis.org> To: mike <mike@coloradosurf.com> Cc: <security@FreeBSD.ORG> Subject: Re: named dying on INSIST Message-ID: <Pine.BSF.4.31.0103271720220.3495-100000@cithaeron.argolis.org> In-Reply-To: <20010326092446.A8530@coloradosurf.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 26 Mar 2001, mike wrote: > I humbly ask the great minds . . . > > Has anyone seen this or can lend insight? > > Over the weekend, my named (primary nameserver) died unexpectedly. > > Mar 24 02:04:53 ns1 named[6470]: > /usr/src/lib/libbind/../../contrib/bind/lib/isc/ev_timers.c:114: > INSIST(now.tv_usec >= 0 && now.tv_usec < 1000000) failed. > Mar 24 02:04:54 ns1 /kernel: pid 6470 (named), uid 53: exited on signal > 6 My machine does this constantly. Are you also getting "microuptime wnet backwards" messages on the consle/syslog? Near as I can tell, my motherboard has an wonky clock chip, which FreeBSD/BIND can't handle. Odd thing is it never had issues until the FreeBSD 4.x series, I can't remember my 3.x machine ever having this problem. There's a few PRs is the GNATS database concerning this. Hopefully your problem will be solved by one of the fixes mentioned. It didn't fix mine. (I have an Open PR on it). Search on "microuptime" will find most of them (some are closed already, so they won't automatically show up in the search). > INSIST(now.tv_usec >= 0 && now.tv_usec < 1000000); This is completely right. The microseconds value of the uptime should never be outside of that range. Unfortunately, all is not well my machine. -- Matt Piechota Finger piechota@emailempire.com for PGP key AOL IM: cithaeron To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-security" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.31.0103271720220.3495-100000>