The default CUPS (Linux print spooler) settings have been shown to be reliable for Eclipse legacy printing at over 100 customer locations. In nearly all instances where print queues are consistently “going down,” a wide area network (WAN) issue has been proven to be the root cause. While working to resolve long-term networking issues, here are a few useful settings that can be adjusted to increase the amount of time allowed by CUPS before disabling a print queue
The Timeout directive controls the amount of time CUPS will wait before an active HTTP or IPP request times out. The default timeout is 300 seconds (5 minutes), but this can be increased by modifying the /etc/cups/cupsd.conf configuration file, per the example below (1200 seconds or 20 minutes):
Maximum Job Limit
The MaxJobs directive controls the maximum number of jobs that are kept in memory. Once the number of jobs reaches the limit, the oldest completed job is automatically purged from the system to make room for the new one. If all of the known jobs are still pending or active then the new job will be rejected.
In a real-world scenario, if there is a printer down with a very large number of jobs in its queue this setting could cause printing to stop functioning server-wide. For this reason, and to ensure there are no hidden hardware or network issues, it is important to periodically monitor the number of jobs in the print queues (using the lpstat or similar tools).
If you regularly submit more than a few hundred jobs at one time to a print queue, you may wish to increase this setting in /etc/cups/cupsd.conf, taking care not to raise it too high to avoid system crash or spooler filesystem from filling up from an out of control print job.
By default, CUPS limits the number of active jobs at any one time to 500, but this can be increased to a larger value or be disabled entirely by setting the value in /etc/cups/cupsd.conf to 0 per the example below: