> The philosophy of the ClamAV team has always been, when in doubt clamd > will not run. There are many people, myself included, who disagree with > this. We have made our objections known, and this is not how the devs > choose to run their project. > > It is their right. I choose to run ClamAV anyway. > > Rant on if it makes you feel better, but you aren't going to change > their view. > > What you SHOULD take from this is that you may want to change how your > milter is set up, so that if clamd dies, unscanned mail is passed > rather than rejected or temp-failed. > > Because if ever a signature database is corrupted, clamd won't run. If > there is an empty signature db file, clamd won't run. There are probably > other situations I can't think of right now that mean clamd won't run. > I think even if the databases are older than a week, clamd won't run.
Mmmh, no: this happened to a client of mine and clamscan kept working. > Now, in the realm of my opinion: If you are running open-source > software then you better be on a mailing list for it, your distro if > not the packages themselves. To my opinion, it wasn't that clear that old installation would have stopped working. Also because it is difficult to find a reason > > Because the Clam team, frankly, owes you nothing. Literally, > absolutely, nothing. They are not only giving you free software, but > daily, hourly, and sometimes MINUTELY updates to the database. Free. > If you stop using it, not only won't it hurt them one bit, it will save > them bandwidth costs. I don't see it this way. I know the team owes us nothing, but a wrong move may hurt them anyway: users may start moving to something else, possibly "smoother" in its reasoning. Other free AV project could stem from this kind of *mistakes*, the ClamAV user-base would decrease and signature updates (which I guess are a function of the virus reported by users) would decrease in effectiveness, too. In summary, the team working on a open-software project has after all some rules to respect in order to keep their own project alive. One of these is: keep the number of users you put in trouble low. Maybe this happened, but I had two calls in the morning about this, for maybe five mailing systems which stopped working. Most of them are not easily upgradeable. After all, I can't care it the less. But what about the five small companies running these systems? > Is it too much to ask that we take the initiative and keep up with what > is available ? If they told us we had to check notes posted on their > front door for updates, it would STILL be more than we deserve. > > I'm sorry, but I literally have no sympathy for people who use something > for free, don't look at announcements for 6 months, then complain things > stopped working. > > And if you think for a minute it would be any better with Microsoft or > Norton or anyone else -- try dealing with a server that just decided > it's now unlicensed for who knows what reason. > > The traffic on this is ridiculous. You don't like it, buy something, > but stop whining already. One of the advantages of free and open software is that one doesn't have to fear for somebody deciding to discontinue service. Even if the ClamAV team decided to stop producing its nice package as well as any cvd update, the problem could have been discovered on the live system (when possible, no hurry) and eventually fixed (again, when possible, no hurry). In this case, instead, the Team proactively stopped any old clamav installation from working. They don't owe anything to their users, but after all even users don't owe their systems to the clamav team... Giampaolo _______________________________________________ Help us build a comprehensive ClamAV guide: visit http://wiki.clamav.net http://www.clamav.net/support/ml
