Hide Forgot
Description of problem: Not sure if this is a issue but I figured I would write it up and let the experts decide. Running instances started from within conductor that exist upon a aeolus-cleanup/aeolus-configure continue to exist in condor's job queue. I am questioning this since the database is wiped of all known instances. Should condor's queue be purged of existing jobs? Not sure why we would allow them to exist since they are essentially orphaned from the app. *Maybe aeolus-cleanup should key off of this and inform the user that instances exist with maybe a option to stop them?* Version-Release number of selected component (if applicable): [root@ce-qe-f14 ~]# rpm -qa | grep aeolus aeolus-configure-2.0.0-4.fc1420110315152531gitf715bd6.noarch aeolus-conductor-0.0.3-0.fc1420110315192501git06dab6c.x86_64 aeolus-conductor-daemons-0.0.3-0.fc1420110315192501git06dab6c.x86_64 aeolus-conductor-doc-0.0.3-0.fc1420110315192501git06dab6c.x86_64 [root@ce-qe-f14 ~]# How reproducible: Steps to Reproduce: 1. start a instance 2. run 'condor_q' from the command line 3. run 'aeolus-cleanup' from the command line 4. run 'aeolus-configure' from the command line 5. run 'condor_q' and see that the job is still there 6. check instances table in the database and see no records Actual results: job still in condor queue Expected results: condor queue purged of existing jobs Additional info: [root@ce-qe-f14 ~]# condor_q -- Submitter: ce-qe-f14 : <10.16.120.173:47516> : ce-qe-f14 ID OWNER SUBMITTED RUN_TIME ST PRI SIZE CMD 3.0 aeolus 3/16 13:15 0+00:11:43 R 0 0.0 job_empty3_3 1 jobs; 0 idle, 1 running, 0 held [root@ce-qe-f14 ~]# psql -U aeolus -d conductor psql (8.4.7) Type "help" for help. conductor=> select * from instances; id | external_key | name | hardware_profile_id | template_id | frontend_realm_id | owner_id | pool_id | provider_account_id | instance_hwp_id | public_addre sses | private_addresses | state | condor_job_id | last_error | lock_version | acc_pending_time | acc_running_time | acc_shutting_down_time | acc_stopped_tim e | time_last_pending | time_last_running | time_last_shutting_down | time_last_stopped | created_at | updated_at | assembly_id | deployment_id ----+--------------+------+---------------------+-------------+-------------------+----------+---------+---------------------+-----------------+------------- -----+-------------------+-------+---------------+------------+--------------+------------------+------------------+------------------------+---------------- --+-------------------+-------------------+-------------------------+-------------------+------------+------------+-------------+--------------- (0 rows) conductor=>
Patch performing the condor_rm on aeolus-cleanup sent to list https://fedorahosted.org/pipermail/aeolus-devel/2011-June/002661.html
condor is out.. closing
closing out old bugs
perm close