[cups.general] Leak in 1.3.7?

Michael Sweet msweet at apple.com
Fri Mar 4 06:10:53 PST 2011


Are you doing raw printing?

I seem to remember an issue in 1.3.x that caused cupsd to leak file descriptors and eventually bring either the system or cupsd to its knees.

On Mar 4, 2011, at 12:57 AM, ALi wrote:

> The leak continues .... we have had a server saturation due to the
> times in cups.
> The saturation of the cups gives the imposibility of login and exec commands
> and we have to restart the machine
> 
> does anybody know if these happens with new versions of cups? or if
> with cups actualization it goes better?
> 
> On Wed, Feb 2, 2011 at 9:58 AM, ALi <osatien at gmail.com> wrote:
>> Hello,
>> we are using cups 1.3.7 (i think it is the default that comes with
>> redhat installation), and we are seeing that there is a delay or leak
>> in the time that takes the lp command to execute.
>> For example we have a time measure in java that measures that the
>> system("lp -d queue file") command and it has started to grow ....
>> from 500 ms to 700 ms more or less, the last time when the measure has
>> grown until 900 the cups failed, so we wont want to wait until that if
>> we can avoid it.
>> We have seen, that when the cups has restarted, the time is improved.
>> So we suspect some kind of leak in the cups, can it be possible?
>> 
>> the system prints 45000 files per day and the delay start to be
>> feeling with a month working ..... from the 350-500 of the start of
>> the month, till the 700-800 of the end of the month
>> 
>> 
>> the cupsd.conf is here http://pastebin.org/HFNmXQ6E (we suspect of
>> something in the cache :S)
>> 
>> thx for all
>> 
> 
> _______________________________________________
> cups mailing list
> cups at easysw.com
> http://lists.easysw.com/mailman/listinfo/cups

________________________________________________________________________
Michael Sweet, Senior Printing System Engineer, PWG Chair





More information about the cups mailing list