Large cups systems?

Seth Galitzer sgsax at ksu.edu
Tue Oct 16 07:16:05 PDT 2007


Kurt Pfeifle wrote:
> Seth Galitzer wrote:
> 
>>> What does 'grep -E "RIPCache|RIP_MAX_CACHE" /var/log/cups/error_log*'
>>> return?
>> D [15/Oct/2007:16:08:32 -0500] [Job 2018] envp[19]="RIP_MAX_CACHE=8m"
>>
>> (repeated a couple of times with different job numbers)
>>
>>> You may be able to get some better performance for larger print jobs
>>> if you set "RIPCache 100m" in your cupsd.conf (instead of the default
>>> value of 8m).
>> What would be an appropriate setting?  Is 100m reasonable for typical
>> jobs, ie jobs that typically are less than 10MB?  I'll have to grep my
>> logs and see what kind of file size I've been processing.
> 
> 
> The RIPCache value limits the RAM used by any CUPS RIP filter running.
> It is mainly there to tell CUPS to be nice to other processes on the
> system, and not take too much resources. If you *do* have the RAM,
> there's no problem setting setting it to a higher value.
> 
> For the default value of RIPCache, I've seen jobs failing to print com-
> plete page images on certain types of raster printers. Setting a higher
> value helped. It also can speed up your printing overall.

Thanks for the info, I went ahead and set it to 100m.

> 
> [....]
> 
>>         lpq command = %p
> 
> That's a bug in Samba. You can work around it by adding a line
> to smb.conf saying
> 
>           lpq command = ""
> 
> (shouldn't have any influence on your problem, though).
> 
I went ahead and added this to my smb.conf, also.

> 
>> I didn't realize those Browse values weren't standard.  They are
>> corrected now.  Yes, this is a VMWare VM.  The host is also Gentoo
>> Linux, running VMWare Server 1.0.3.44356.  This guest VM has a 10GB
>> (3.8GB free) virtual disk and 2GB RAM. 
> 
> That doesn't matter at all if the hosting OS machine (the physical
> one) has little or no RAM/disk space left.
> 
> So what do these two commands give back on your *host* Gentoo?
> 
>   df -h
>   cat /proc/meminfo

The volume I have the VM disk on has 21GB free, and the machine has 6GB
RAM total.  I only have one other guest VM running on there with 1GB of
RAM allocated

> 
> ?
> 
> You should indeed grab the next few spoolfiles which provoke a "hanging
> job" and save them to somewhere else, and note down the target printers,
> the job settings used, etc.
> 
> (The files will be in the original format as received by CUPS, prior to
> any filtering being applied).
> 
> Then you can use these to try and reproduce the problem. You can also
> see if the files would print if you sent them to a different printer
> (model).
> 

Thanks for the tip.

I can tell you that the couple of changes I made last night to
cupsd.conf (sane Browse values and adding ErrorPolicy) seem to have made
a big difference.  When I looked this morning, I only had one queue that
was stalled, and it was one I had tried using the BackEnd Handler on.
Now that I know the ErrorPolicy/JobRetry settings basically perform the
same function, I'll just use those instead.

I'll keep watching things today and let you know how it's going.

Thanks for your extensive help.

Seth




More information about the cups mailing list