[Gllug] CUPS

[rdrinkwater]

Morning all.

We are running Mandrake 8 and are using CUPS.  A number of printers are 
attached via a mix of print servers, serial ports and terminal servers. 

We are experiencing strange print problems whereby the CUPS daemon 
suddenly "grabs" huge blocks of memory (> 100Mb) and tops reports between 
90% and 97% proc usage.  At the same time all print output ceases.  This 
condition appears to be triggered either when the number of print jobs in 
/var/spool/cups is large (such as when statements are printed) or when a 
printer is made off-line during one of the volume print runs (because it 
runs out of paper).

Usually after a period of time ( anything between 40 and 120 minutes) 
printing restarts.  Sometimes the only way to restart printing is to kill 
the CUPS daemon and restart it.

A further complication is that during the "print lockup" the lpstat -t 
command hangs (ctrl\ to quit) aa does the enable/disable printer command.

I have reported this problem to cups.org (black hole job!) and posted a 
report on the support bulletin (as yet no replies).

Has anyone experienced similar problems?  If so have you found a 
remedy(ies)?

All thoughts on how to progress this (perhaps ways to find out what CUPS 
is hanging on etc) will be gratefully recieved.  BTW CUPS operates 
through http://localhost:603 port.

Roy Drinkwater.


-- 
Gllug mailing list  -  Gllug at linux.co.uk
http://list.ftech.net/mailman/listinfo/gllug