Description of problem: The init script's stop function always sleeps for 5 seconds, which is often an arbitrary delay Version-Release number of selected component (if applicable): 7.2.0-3 How reproducible: 100% Steps to Reproduce: 1. service condor start 2. service condor stop Actual results: 5 second delay Expected results: Delay as long as it takes to stop condor Additional info:
This will be fixed in 7.2.0-4 commit ca6199060ceff5d5008bdfdf46de2a483d3d8dc0 Author: Matthew Farrellee <matt> Date: Wed Jan 21 09:47:35 2009 -0600 Fixed silly sleep 5 in init stop It was replaced with a call to wait_pid, which waits up to 15 seconds for the condor_master to exit. An existing issue is if the master still hasn't exited after those 15 seconds and stop was invoked as part of restart (or if start is invoked before the condor_master exists, generally) the subsequent start will think that the master is already running, which it is, and will not do anything. The result is the master will eventually exit and not actually be restart.
The bug in $SUBJ is fixed (see below). I have verified on RHEL5.3 i386, condor-7.2.1-0.1.el5. It just seems that on freshly installed condor, there is no directory set-up to store the logs as a non-privileged user: --------------------------------------------------- rhel5:~# /etc/init.d/condor start Starting Condor daemons: Can't open "/var/log/condor/MasterLog" dprintf() had a fatal error in pid 3315 Can't open "/var/log/condor/MasterLog" errno: 2 (No such file or directory) euid: 64, ruid: 0 [FAILED] rhel5:~# mkdir /var/log/condor rhel5:~# /etc/init.d/condor start Starting Condor daemons: Can't open "/var/log/condor/MasterLog" dprintf() had a fatal error in pid 3328 Can't open "/var/log/condor/MasterLog" errno: 13 (Permission denied) euid: 64, ruid: 0 [FAILED] rhel5:~# chgrp condor /var/log/condor rhel5:~# chmod g+w /var/log/condor/ rhel5:~# /etc/init.d/condor start Starting Condor daemons: [ OK ] rhel5:~# service condor stop Stopping Condor daemons: [ OK ] rhel5:~# /etc/init.d/condor start Starting Condor daemons: [ OK ] rhel5:~# time service condor stop Stopping Condor daemons: [ OK ] real 0m4.291s user 0m0.080s sys 0m0.087s rhel5:~# /etc/init.d/condor start Starting Condor daemons: [ OK ] rhel5:~# time service condor stop Stopping Condor daemons: [ OK ] real 0m5.199s user 0m0.080s sys 0m0.077s rhel5:~# /etc/init.d/condor start Starting Condor daemons: [ OK ] rhel5:~# time service condor stop Stopping Condor daemons: [ OK ] real 0m6.262s user 0m0.097s sys 0m0.090s rhel5:~# /etc/init.d/condor start Starting Condor daemons: [ OK ] rhel5:~# time service condor stop Stopping Condor daemons: [ OK ] real 0m1.131s user 0m0.070s sys 0m0.053s ---------------------------------------------------
The log/condor directory problem is being solved in https://bugzilla.redhat.com/show_bug.cgi?id=476892
An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on therefore solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHEA-2009-0434.html