> Well, I can prove the failure for > root # /usr/clamav/sbin/clamd --version > clamd / ClamAV version 20030806 > selfbuilt with exim/exiscan 4.21. > one thread goes defunct, the master thread idles around and the third is > stuck in an empty event loop. > Exim connections pile up to (in this case: 777) connections and the mail > exchanger grinds down. > After killing clamd > stop) > pkill -9 -P 1 clamd && sleep 1 && echo 'Clam anti virus daemon stopped' > and waiting some additional time(!!!) it can be restarted > start) > rm -f /var/clamav/run/clamd > clamd && echo 'Clam anti virus daemon started' > Note the manual deletion of the socket. Bad bad bad bunny!
Yes, this is all exactly what we experience. > What exactly do You mean with "more robust against mail bodies"? There > should be no way to break the scanner with "bad mail". It should *protect* > other systems, not being exploitable(?) itself. Theoretically your right, but it does take time to develope software that meets that high criteria. CLAMAV is relatively young. > So, which version *is* stable? Apparently, for the milter work, none. It probably will be some day, for now we'll go find something else and check back later (since 20030806 isn't stable either). ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ Clamav-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/clamav-users
