[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#899259: [apple/cups] systemd socket activation and printing (#5319)



On Mon 04 Jun 2018 at 16:46:49 +0000, Michael R Sweet wrote:

> OK, so the issue is that, after printing something, that cupsd does
> not idle exit after 60 seconds?

That about sums it up; the behavior is in contrast to using lpadmin,
lpstat etc and getting idle exit after 60 seconds

> Keep in mind that what cupsd considers idle may not be your
> expectation - if printer sharing or the web interface is enabled, if a
> job is still being processed, or other child processes are still
> active, then cupsd will not idle exit (by design).

Speaking for my own tests: the web interface was not enabled, there were
no jobs being processed and no printer sharing. Indeed, for the lpadmin
command the log shows

> Printer sharing is off and there are no jobs pending, will restart on demand.            
> Scheduler shutting down normally.

Child processes? I wouldn't know what I was looking for.

> Anyways, from the limited log snippets in the Debian bug report, it
> looks like select_timeout() is returning 86400 (one day) because there
> is nothing to do and we don't wake up just to write out dirty files. I
> can look at updating the code to reduce the timeout when we would
> normally idle exit, but (as mentioned in my recent OpenPrinting
> presentation) doing an idle exit isn't always the optimal solution
> anyways.

FTR:

https://ftp.pwg.org/pub/pwg/liaison/openprinting/presentations/cups-plenary-may-18.pdf

If a fuller log is required, I'm sure Francesco Poli would provide one.

Regards,

Brian.


Reply to: