From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.0.1) Gecko/20021003 Description of problem: When CUPS is started right after a Phoebe2 install for the first time it takes ages. On my Athlon XP 2000+ it took over 3 minutes to come. This could make Joe User nervous... When booting the system again CUPS start in seconds. Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1. Install Phoebe2 2. Watch how long CUPS needs to start for the first time 3. Additional info:
I noticed this as well. Installed phoebe 2 on 5 machines around 1GHz in speed and the cups startup -- on the initial boot -- took several minutes.
Do you both have gimp-print-cups installed?
I do. This is an "Everything" install of Phoebe2.
Moved this processing time into the gimp-print-cups post-install scriptlet, where it should have been to start with. :-) gimp-print-cups-4.2.4-5 cups-1.1.17-10
I have this problem every time I (re)start cups: [...] root@wombat:~> service cups restart Stopping cups: [ OK ] Starting cups: [ OK ] root@wombat:~> rpm -q gimp-print-cups gimp-print-cups-4.2.4-5 root@wombat:~> rpm -q gimp-print-cups cups gimp-print-cups-4.2.4-5 cups-1.1.17-13 root@wombat:~> service cups restart Stopping cups: [ OK ] Starting cups: [ OK ] root@wombat:~> time service cups restart Stopping cups: [ OK ] Starting cups: [ OK ] real 0m40.890s user 0m0.900s sys 0m0.320s root@wombat:~> time service cups restart; dmesg|grep -v FILTER Stopping cups: [ OK ] Starting cups: [ OK ] real 0m40.936s user 0m0.820s sys 0m0.360s 0x00 PREC=0x00 TTL=124 ID=40592 DF PROTO=TCP SPT=2410 DPT=4662 WINDOW=32767 RES=0x00 SYN URGP=0 usb_control/bulk_msg: timeout usb_control/bulk_msg: timeout usb_control/bulk_msg: timeout usb_control/bulk_msg: timeout usb_control/bulk_msg: timeout usb_control/bulk_msg: timeout usb_control/bulk_msg: timeout usb_control/bulk_msg: timeout root@wombat:~> [...] I take it that 40 seconds is a bit harsh for starting cups when all I have is a LaserJet 5L attached to the parallel port (so nothing fancy here). I see that there are a lot of USB timeouts here (the "grep -v FILTER" part is to leave out (most/complete) messages from iptables) -- is there a means of telling CUPS not to try accessing USB printers (or do it in the background or something similar)?`
Nils: this is a separate problem, and I think it warrants a separate report. Could you also include details of USB chipset (at least which module is used)? Arjan has a USB printer that seems to trigger this 'timeout' behaviour, although CUPS doesn't take any longer for him than normal to start. Arguably a kernel issue really.