Hi Mike,
Look in your postgresql log file (probably in /var/log/postgresql) for postgres
errors. And you probably want to restart postgres.
A quick test from psql would be the getrunnable select below:
> SELECT * FROM getrunnable() LIMIT 10;
Bob Gobeille
On Aug 22, 2011, at 12:10 PM, Fay Michael T wrote:
> An update, I restarted everything and now I am seeing the following errors to
> the console (but not in the logs):
>
> # *** Exiting fossology-scheduler PID 10266 ***
> Exec failure reason: Bad file descriptor
> Exec failure reason: Bad file descriptor
> Exec failure reason: Bad file descriptor
> Exec failure reason: Bad file descriptor
> Exec failure reason: Bad file descriptor
> Exec failure reason: Bad file descriptor
> ERROR: DBaccess(7):
> ERROR: DBaccess error: 'SELECT agent_pk FROM agent WHERE agent_name =
> 'unpack' AND agent_rev = 'default';'
> ERROR: DBaccess(7):
> ERROR: DBaccess error: 'INSERT INTO agent (agent_name,agent_desc,agent_rev)
> VALUES ('default','default agent for use with scheduler','default');'
> ERROR: DBaccess(7):
> ERROR: DBaccess error: 'SELECT currval('agent_agent_pk_seq'::regclass);'
> Exec failure reason: Bad file descriptor
> Exec failure reason: Bad file descriptor
> Exec failure reason: Bad file descriptor
> Exec failure reason: Bad file descriptor
> Exec failure reason: Bad file descriptor
> Exec failure reason: Bad file descriptor
> ERROR: DBaccess(7):
> ERROR: DBaccess error: 'SELECT * FROM getrunnable() LIMIT 10;'
> ERROR: DBaccess(7):
> ERROR: DBaccess error: 'SELECT DISTINCT(jobqueue.*), job.* FROM jobqueue LEFT
> JOIN jobdepends ON jobqueue.jq_pk = jobdepends.jdep_jq_fk LEFT JOIN jobqueue
> AS depends ON depends.jq_pk = jobdepends.jdep_jq_depends_fk LEFT JOIN job ON
> jobqueue.jq_job_fk = job.job_pk WHERE jobqueue.jq_starttime IS NULL AND (
> (depends.jq_endtime IS NOT NULL AND depends.jq_end_bits < 2 ) OR
> jobdepends.jdep_jq_depends_fk IS NULL) ORDER BY job.job_priority
> DESC,job.job_queued ASC LIMIT 6;'
> ERROR: DBaccess(7):
> ERROR: DBaccess error: 'UPDATE scheduler_status SET agent_status='RUNNING',
> agent_status_date='now()', record_update=now(), agent_param=E'' WHERE
> unique_scheduler='ldxapp27.dx.deere.com.21299' AND agent_number='-1';'
>
>
>
>
> From: fossology-boun...@fossology.org
> [mailto:fossology-boun...@fossology.org] On Behalf Of Fay Michael T
> Sent: Monday, August 22, 2011 12:18 PM
> To: fossology@fossology.org
> Subject: [FOSSology] FW: Fossology
>
> For some reason, when we submit tasks, they won’t start. Everything is in a
> pending status and I can’t even get the scheduler through the admin function
> in the web interface to load.
>
> I had to restart the scheduler today because it was exhibiting some kind of
> race condition where the watchdog got stuck continually restarting a
> non-responding scheduler, until it simply couldn’t open a new scheduler
> because it had too many files open.
>
> 2011-08-22 09:49:43 scheduler[7485] : *** failed to open lock file for
> fossology-scheduler (see LockGetPID). Too many open files
> 2011-08-22 09:50:45 scheduler[7485] : *** Scheduler restarted successfully by
> fo_watchdog ***
> 2011-08-22 10:10:20 scheduler[7485] : *** Scheduler not responding: killing
> and restarting ***
> 2011-08-22 10:10:20 scheduler[7485] : *** failed to open lock file for
> fossology-scheduler (see LockGetPID). Too many open files
> 2011-08-22 10:11:24 scheduler[7485] : *** Scheduler restarted successfully by
> fo_watchdog ***
> 2011-08-22 10:40:18 scheduler[7485] : *** Scheduler not responding: killing
> and restarting ***
> 2011-08-22 10:40:18 scheduler[7485] : *** failed to open lock file for
> fossology-scheduler (see LockGetPID). Too many open files
> 2011-08-22 10:41:26 scheduler[7485] : *** Scheduler restarted successfully by
> fo_watchdog ***
> 2011-08-22 11:08:16 scheduler[7485] : *** Scheduler not responding: killing
> and restarting ***
> 2011-08-22 11:08:16 scheduler[7485] : *** failed to open lock file for
> fossology-scheduler (see LockGetPID). Too many open files
> 2011-08-22 11:09:38 scheduler[7485] : *** Scheduler restarted successfully by
> fo_watchdog ***
>
> Any ideas where I should be looking to figure out what’s happening.
>
> This is 1.4 running on RHEL 5.
>
> From: Christopher Terry D
> Sent: Monday, August 22, 2011 11:01 AM
> To: Fay Michael T
> Subject: RE: Fossology
>
> Everything is pending
> <image001.png>
>
> From: Fay Michael T
> Sent: Monday, August 22, 2011 10:57 AM
> To: Christopher Terry D
> Subject: RE: Fossology
>
> Did prod ever complete?
>
> From: Christopher Terry D
> Sent: Thursday, August 18, 2011 11:05 AM
> To: Fay Michael T
> Subject: Fossology
>
> I ran a few things through both test and prod of Fossology. All the test
> stuff came back and I got what I needed done. The Prod still hasn’t come back
> with a completed email yet. Last I looked it was hung up again…
>
> T.C. - Terrence Christopher
> Project Manager, Open Source Governance
> Deere & Company World Headquarters
> One John Deere Place, Moline IL 61265
> Office: (309) 765-4368
> Mobile: (309) 716-4998
>
> <ATT00001..txt>
_______________________________________________
fossology mailing list
fossology@fossology.org
http://fossology.org/mailman/listinfo/fossology
