Date: Sun, 05 Jan 1997 17:02:28 -0600 From: dkelly@hiwaay.net To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) Cc: current@FreeBSD.ORG, randyd@nconnect.net (Randy DuCharme) Subject: Re: syslogd failure Message-ID: <199701052302.RAA16060@nexgen.HiWAAY.net> In-Reply-To: Message from j@uriah.heep.sax.de (J Wunsch) of "Sat, 04 Jan 1997 11:34:28 %2B0100." <Mutt.19970104113428.j@uriah.heep.sax.de>
next in thread | previous in thread | raw e-mail | index | archive | help
joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) replied: > > As dkelly@HiWAAY.net wrote: > > > Don't know why, but "rm /var/run/log" and restarting syslogd fixed the > > problem. Did my filesystem get dirty somehow? > > No, that's already okay. It's normally done at /etc/rc time. fsck or "rm /var/run/log" at /etc/rc time? (Sorry, I can't get to my -current system right now from here.) The problem persisted thru "shutdown -r now". My (recent) thought was the filesystem was somehow hosed but the dirty bit was cleared because I did a proper shutdown. > The question is however _why_ syslogd dies in the first place. If you > can afford it, i'd suggest to do the following: > > . kill the default syslogd > . rm /var/run/log > . switch to an otherwise unused VTY, where it can run undisturbed > . start it there with -d, this will make it spit out some more > messages, and run in the foreground > . once it died, try to find whether you're seeing something on that > VTY (do also watch out for the exit status before clobbering it!) I've filed this message for future reference. Will do the above if the problem arises again. The problem was with a late October -current. Logging to /var/log/messages stopped in mid- to late- November. Was getting these warnings about syslogd but wrote it off as, "what do you expect when running -current?" After getting syslogd running again I never had the problem again. Have since run "make world" a couple of times... So unless it pops up again, I've lost the opportunity to track it down. Should mention there were other things happening on this system, such as the difference between write-thru and write-back and none, for CPU cache on an el-cheapo VIP 486/5x86 PCI/VL MB. Its sorta hard to debug more than one thing at a time. -- David Kelly N4HHE, dkelly@hiwaay.net ===================================================================== The human mind ordinarily operates at only ten percent of its capacity -- the rest is overhead for the operating system.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199701052302.RAA16060>