[cups.development] Writing Jobs Files to Database Not Files

Michael Sweet msweet at apple.com
Thu Apr 5 10:19:06 PDT 2012


One of the issues of this approach is that you have a single-point-of-failure in the database. To avoid this you can instead have the database system query each of the individual servers to get the job history (perhaps using IPP notifications to know when to fetch more job information...)

Similarly, you can use a custom backend on the servers to implement any quota system you like and locally cache the most recent network-wide totals maintained by the database server so that a temporary outage of any one server does not impact normal usage.


On Apr 4, 2012, at 1:36 AM, Michael Hull <mhull at hull-innovations.com.au> wrote:

> Hello,
> 
> I am new to coding and development. I have a specific goal in mind to create an accounting system with CUPS. One of the first things I want to do is have the "Jobs Files" and "Log Files" (In the high level diagram) logging to a database rather than the file system.
> 
> I can't see it too difficult to add the jobs as blobs to a database but I am concerned about performance, will adding another data retrieval layer adversely affect CUPS performance?
> 
> I want to do it this way for clustering purposes, rather than clustering CUPS software (Which I realise that I will have to do anyway if your using more than one print server on a site), I think it would be easier to run a database cluster and have multiple independent CUPS front ends. (Hit print on one site, collect on another).
> 
> I would appreciate and feedback about this approach and any tips anyone may have.
> 
> Thanks,
> 
> Michael Hull
> Hull Innovations
> 
> 
> _______________________________________________
> cups-dev mailing list
> cups-dev at easysw.com
> http://lists.easysw.com/mailman/listinfo/cups-dev

__________________________________________________
Michael Sweet, Senior Printing System Engineer, PWG Chair





More information about the cups mailing list