cupsd hang condition

Kelly Sauke ksauke at fastenal.com
Fri May 27 07:07:42 PDT 2005


I'm running 1.1.23 on RHEL 3 compiled myself.  Here's the bug.

I have a client which does a BrowsePoll to 2 servers that both have an identical list of printers on them.  I print a bunch of jobs to the ImplicitClass that gets generated from
the two servers.  Now say one print server crashes or becomes unreachable.  If there are jobs queued to the ImplicitClass when the client hits the BrowseTimeout on the crashed
print server, it will lock up and spike the CPU.  You have to kill -9 the cupsd process in order to get it die.  The last thing in the logs is usually "Remove destination
"printer at server" has timed out; deleting it...".  I've been able to reproduce this a couple of times now.  If you need further information let me know and I'll provide whatever you
need.




More information about the cups mailing list