I have two servers on the same subnet, one CUPS server and one that needs to access printers configured on the CUPS server. CUPS browsing seems to be partially working, but according to lpstat on the client, the devices are
///dev/null instead of an
[username@clienthost ~]$ lpstat -t scheduler is running no system default destination device for boardroom: ipp://oldprintserver.my.network:631/printers/boardroom device for br: ipp://oldprintserver.my.network:631/printers/br device for its: ///dev/null <------ THIS IS BAD?
- Print server is running Scientific Linux 6.4 with CUPS 1.4.2
- Printing client is running CentOS 5.8 with CUPS 1.3.7
Server configuration - cupsd.conf
Browsing On BrowseRemoteProtocols BrowseAddress @LOCAL BrowseLocalProtocols CUPS dnssd
Client configuration - cupsd.conf
Browsing On BrowseOrder allow,deny BrowseAllow @LOCAL BrowsePoll therealprintservername.my.network BrowseInterval 30
Using tcpdump to analyze packets between hosts, it's clear that the client is polling the server over IPP. The server responds with 200 OK. Also, it seems that the client is only seeing the classes advertised by the print server, not the printer queues themselves.
So here's what I really want to know.
- What is the minimum required configuration for CUPS browsing to work on the same LAN?
- Does printer browsing work between different versions of CUPS?
- What other troubleshooting tools are available for problems with printer browsing?
[Edit: Even though lpstat -t says the device is ///dev/null, I can still send print jobs to the device. So it looks like basic printing works, but I'm still wondering why it shows as ///dev/null on the client.]