On Tue, Jun 29, 2004 at 08:32:08PM +1000, Robert S said: > I don't believe that there are any milter processes after the init script > exits. I have written init scripts in the past for slackware and found that > you need to pause before reloading the milter after doing a "killall" > (usually about 3 seconds is OK) - I guess what you describe is a similar > thing. Maybe I'll just revert back to the default init script - it sounds > safer (even if it is a bit ugly). Its much less likely to break when I > upgrade the clamav package.
Good - I'm glad the init scripts work for you as they do for me - I don't have too many sendmail environments to test on, and the ones I do have are all heavily loaded machines (hence the init script being written that way). > My mail server is only going to be for a small office - I don't think that > "heavily loaded" applies here. Would it be feasible to change the 30-second > timeout, or should I just leave well alone? The easiest from an admin point of view is leave well enough alone for now. I am trying make a point of reviewing the init scripts and other package parts every new version or so, and seeing if old hacks are still needed. I've pulled out several in the 0.7x series of clam, so maybe in 0.8x we can get rid fo this one as well. Sorry it's ugly, and I will be as happy as you to see that piece of code go, -- -------------------------------------------------------------------------- | Stephen Gran | The sweeter the apple, the blacker the | | [EMAIL PROTECTED] | core -- Scratch a lover and find a foe! | | http://www.lobefin.net/~steve | -- Dorothy Parker, "Ballad of a Great | | | Weariness" | --------------------------------------------------------------------------
pgp4wUWkhEYjK.pgp
Description: PGP signature
