> These are the entries I copied directly from the newly-installed > startup scripts in /usr/local/etc/rc.d, over to /etc/rc.conf, then > simply > executed the startup scripts without so much as a reboot. I must admit, > though, that when faced with any balky startup script difficulties with > installations in general, having used FreeBSD since /etc/sysconfig days, > I'm prone to resorting to a plain-old-fashioned rc.local entry. :) > > If you're having trouble getting things to start, make sure you enable > every logging and debug knob you can find, then check the logs, after > your start-up attempt.
I have three of these machines out in a round robin setup, all running fbsd 5.2.1 and clamav 0.75. I built a 5.3 machine with .80 and threw it into the mix to see what results I would get, as testing wasn't telling me anything (debugging and logging). The only real difference that I saw was that the conf file changed names and such. However after any sort of real stress hit the machine, clamd would die and that particular server would reject mail. Needless to say, he came out of the mix rather quickly. I have been looking for reasons as to why that could have been happening. I know what you mean in regard to using rc.local as opposed to the "newish" startup script method when there seem to be problems. I even tried a similar method when I was initially troubleshooting the sporadic death of clamd, but to no good end. I eventually found that there were all sorts of people having the same issue with multiple versions of clamav. A little disconcerting that the best solution I have seen so far is the "cron job/ script to see if the daemon is running". I was using that as a last gasp effort since I didn't want to reject mail unnecessarily. I haven't been on the list very long though, and there's a lot to be said for 93 days without a glitch ;) Thx. _______________________________________________ http://lists.clamav.net/cgi-bin/mailman/listinfo/clamav-users
