Hide Forgot
Description of problem: I am not sure exactly how to reproduce the wedging, though I currently have a system where it is occurring. In any case, I noticed that my template promotions weren't taking place, so I checked status of the katello-jobs status. Boiled down, it comes to this: [root@se-rhelbox katello]# service katello-jobs status katello-jobs is not running. [root@se-rhelbox katello]# service katello-jobs start; cat jobs-startup.log Starting katello-jobs: [ OK ] /usr/share/katello/app/models/sync_plan.rb:35: warning: don't put space before argument parentheses ERROR: there is already one or more instance(s) of the program running So there are two issues, probably related: 1) the service script thinks something is not running (when it may be?) 2) Somehow it is returning an OK on the start, even though the error logs indicate otherwise. Also possibly useful info: [root@se-rhelbox katello]# ps ax|grep job 14431 ? S 0:01 katello/delayed_job 14482 ? S 0:01 katello/delayed_job 14561 pts/3 S+ 0:00 grep job 19873 ? S 0:01 delayed_job_monitor
[root@se-rhelbox katello]# service katello-jobs start Starting katello-jobs: [ OK ] [root@se-rhelbox katello]# echo $? 0
It's due to very poor daemon interface of delayed_jobs. Will pick this up later on.
taking
https://github.com/Katello/katello/pull/613
| * 0e78e3f (fork/pull-req-bz754738, pull-req-bz754738) 754738 - do not override variables in other procedures | * 804d5ba 754738 - do not override status() from /etc/rc.d/init.d/functions | * 73c80bf 754738 - fix name of monitor pid file | * 2a98a6c 754738 - if program is already running, print failure, but return 0 | * ba2f8c6 754738 - if we fail in stopping delayed_jobs, kill it. One by one. | * 225d592 75473 - correctly solve status for all processes of delayed_jobs | * 2bd03eb 754738 - log even output of service stop | * ddd5d85 use runuser instead of su | * f3e3106 75473 - do not delete nor truncate log | * 6048c1a 754738 - properly return when katello is not configured
Part of Katello 1.1 release