Description of problem: Using the web admin menu clicking on the administration tab causes 100% cpu and an extended delay (minutes not seconds) Version-Release number of selected component (if applicable): cups-1.2.0-1.1 How reproducible: Slow every time Steps to Reproduce: 1.Open web browser 2.localhost:631 3.click on administration and wait.... Actual results: Works eventually Expected results: Works quickly Additional info:
What processor are you using, and what is it's clock speed? Also, can you attach /var/log/cups/error_log? Thanks.
I have tried multiple machines, my personal machine is a dual core amd 4400 I [26/May/2006:09:05:46 +1000] Full reload complete. I [26/May/2006:09:05:46 +1000] Listening to :::631 on fd 2... I [26/May/2006:09:05:46 +1000] Listening to 0.0.0.0:631 on fd 3... E [26/May/2006:09:05:46 +1000] Unable to set ACLs on root certificate "/var/run/cups/certs/0" - Operation not supported I [26/May/2006:09:05:52 +1000] Started "/usr/lib/cups/cgi-bin/admin.cgi" (pid=14412) I [26/May/2006:09:05:52 +1000] Started "/usr/lib/cups/daemon/cups-deviced" (pid=14413) I [26/May/2006:09:06:22 +1000] cupsdCloseClient: SSL shutdown successful! W [26/May/2006:09:06:22 +1000] [cups-deviced] Backend "snmp" did not respond within 30 seconds! I [26/May/2006:09:06:24 +1000] Started "/usr/lib/cups/cgi-bin/admin.cgi" (pid=14436) I [26/May/2006:09:06:24 +1000] Started "/usr/lib/cups/daemon/cups-deviced" (pid=14437) W [26/May/2006:09:06:55 +1000] [cups-deviced] Backend "snmp" did not respond within 30 seconds! E [26/May/2006:09:09:16 +1000] PID 14412 (/usr/lib/cups/cgi-bin/admin.cgi) crashed on signal 9! I [26/May/2006:09:09:16 +1000] Hint: Try setting the LogLevel to "debug" to find out more. I [26/May/2006:09:10:57 +1000] cupsdCloseClient: SSL shutdown successful! I [26/May/2006:09:10:57 +1000] cupsdCloseClient: SSL shutdown successful!
Please run this command: strace -t /usr/lib/cups/backend/snmp 2>snmp.log and attach the snmp.log file. Thanks.
Created attachment 130164 [details] snmp log as requested
We have tracked it down from that log to a particular printer. A Kyocera FS1010 with a nic. When we turn it off the admin page loads much faster. I would not mind if you closed this call. Cheers
Thanks. It's something that would be very good to fix, so I'll keep this report open.
The same effect on freshly installed FC5, updated with old HP Laserjet 4 connected thriugh lpt port. On IBM Thinkpad t41 with fc5 with the same config everything works OK.
Reported upstream: http://cups.org/str.php?L1737
Could you please run this command: CUPS_DEBUG_LEVEL=3 /usr/lib/cups/backend/snmp 2>&1 >snmp.log and attach the snmp.log file here? Thanks.
Created attachment 130593 [details] updated snmp log as requested
Created attachment 130663 [details] snmp log
Please try this test update: https://www.redhat.com/archives/fedora-test-list/2006-July/msg00666.html You should be able to install this using the command: yum --enablerepo=updates-testing 'cups*' as root. Does the backend still take a long time? If so, please show me the snmp.log from: CUPS_DEBUG_LEVEL=3 /usr/lib/cups/backend/snmp 2>&1 >snmp.log with the new (1.2.2-1.2) packages installed. Thanks.
No, backend seems to work ok. The admin page shows after a short time with proper content. Thnx