ANNOUNCE: Unstable 4.30.1

[Raymond Dijkxhoorn]

Hi!

> > - Fixed problem with some systems not rewinding file extraction directory
> >    properly.
> > - Fix to get rid of zombie processes quickly and cleanly.
> 
> I dont know if it has to do with the zombie code, but i have it running on 
> two of my exim boxes currently, 4.30.1 and both show the same behaviour. I 
> have 10 mailscanner children normally, but they just die after some time 
> and wont come back. Its all running ok, for hours, but then they drop out 
> and i have to restart mailscanner to get it going again. Noticed this on 
> my 2nd box, and seeing this on my 1st one also now.

In addition to this, also get this on the one other box. With other linux 
version. So this is happening with both RH9 and Fedora CORE-1. 

[root at vmx50 root]# pstree
init-+-MailScanner---MailScanner

Where i get this after a restart:

[root at vmx50 root]# pstree
init-+-MailScanner---14*[MailScanner]

Since normally 14 are running, so thats ok. 

And alos there:

Apr 16 22:56:04 vmx50 MailScanner:  succeeded
Apr 16 22:56:25 vmx50 application bug: MailScanner(7572) has SIGCHLD set 
to SIG_IGN but calls wait().
Apr 16 22:57:25 vmx50 application bug: MailScanner(9744) has SIGCHLD set 
to SIG_IGN but calls wait().
Apr 16 22:58:25 vmx50 application bug: MailScanner(11175) has SIGCHLD set 
to SIG_IGN but calls wait().
Apr 16 22:59:26 vmx50 application bug: MailScanner(10800) has SIGCHLD set 
to SIG_IGN but calls wait().

I have this on all 4 boxes i am running the beta on, 3 are running with 
exim, one with sendmail.

Looks like some structural problem in the new version, anyone noticed, or 
am i the only one running it :) 

Bye,
Raymond.