[hylafax-users] zombie process

["Alain Wenmaekers" <spicyalan@pandora.be>]

Hi everyone,

 

I have a Hylafax 4.1beta3 faxserver (an 486 running RH 7.0) with a Rockwell modem (in class2...did work on default detection). It's all working perfectly, except that I see this coming quite frequently:

 

uucp     17619     1  0 13:41 ?        00:00:00 faxq
uucp     17629     1  0 13:41 ?        00:00:00 hfaxd -i hylafax
uucp     18178 17629  0 18:37 ?        00:00:11 [hfaxd <defunct>]

 

appearantly there is a client somewhere that disconnects wrong (all running the most recent cypheus on Windows2000). I can't really see which one (too many people around to check this). actually most of them keep there PC and cypheus client running 24/24.

 

If I see it, I do a /etc/rc.d/init.d/hylafax restart, and then it is back normal..... But can this do any arm?....I would like it to go away after some time if this happens. Restarting the hylafax is a problem (you'll never know if someone begins faxing...and also everyone gets disconnected).

 

I'm going to try this weekend to keep it running with a defunct process....maybe it goes away by itself.

 

It is not really a big problem, because all the other processes continues to run perfectly (so I can live with it :-)

 

Can anyone give more info on this...or has anyone have seen something similar?

 

 

Grtz

 

Alain