On 12-01-06 08:59 AM, Jason R Peak wrote: > Hi Greg, > Rather than risk rushing through your suggestions at the end of my workday, I > have saved them for this morning and a fresher mind. > Unfortunately, I am not seeing improvement: > The scheduler service is still not listed in my core MH welcome page, and the > schedule.ics errors remain in the logs. > > Having figured out most of the configuration settings, it seems as though I > could shift my efforts towards a fresh installation of 1.3. > Any cautions against this? I could help test, and the troubleshooting for 1.3 > is certainly fresher in everyone else's mind...
Is this for production use? I would be leery of using an unreleased version in production, mainly because you will run into bugs that other adopters have not (yet). If this is just proof of concept or a pilot then yeah for sure. Switching between versions is as easy as a new checkout of the code and rebuilding the 3rd party tools. > See below for the results of my latest efforts with 1.2. > > Many thanks, > Jason > > >> Can you try it using the autogenerated >> tables? Wipe the DB, and turn on org.opencastproject.db.ddl.generation >> in config.properties. > I may not have wiped the DB as completely as I needed to for > org.opencastproject.db.ddl.generation to do its thing. > I manually emptied the tables via the mysql client SQL command DELETE FROM > <TABLE_NAME>; for each table in the db. > For the DICTIONARY table, DELETE FROM... seemed like it would take several > hours to complete. Impatient, I simply dropped the table (which only took > 2min 45sec!). > On the next start of Matterhorn, the DICTIONARY table did not get recreated, > although I thought that it would as a result of > org.opencastproject.db.ddl.generation being set to 'true'. Hrm, that table should be created more or less on startup. When I clear the DB I issue the following command: drop schema matterhorn; create schema matterhorn; Short, quick, and completely wipes out everything Matterhorn related in the db. >> It looks to me like some of the bundles aren't starting correctly on >> both your core and your CA, so let's try rebuilding them and making sure >> that the build finishes successfully. > > Perhaps it's the fresh restart of Core that is turning up these > com.mysql.jdbc.exceptions.jdbc4.MySQLSyntaxErrorException events (I have seen > these before). > Otherwise, the ERRORs and WARNs are the same as before the fresh rebuild. Ah, there we go. The tables that the code are expecting and the tables its finding are different (there's a missing column at the very least). Try wiping out the DB (as above) and restarting. This should recreate the tables as each module starts and hopefully get you a working system. G > CORE: > ================== > 2012-01-06 08:47:36 WARN (SchedulerImpl:599) - Invalid calendar data, > skipping parse attempt. > 2012-01-06 08:47:36 WARN (IoSupport:241) - IOException attempting to get > file from file:/var/spool/opencast/cache/schedule.ics. > 2012-01-06 08:42:36 WARN (SchedulerImpl:599) - Invalid calendar data, > skipping parse attempt. > 2012-01-06 08:42:36 WARN (IoSupport:241) - IOException attempting to get > file from file:/var/spool/opencast/cache/schedule.ics. > 2012-01-06 08:37:36 WARN (SchedulerImpl:599) - Invalid calendar data, > skipping parse attempt. > 2012-01-06 08:37:36 WARN (IoSupport:241) - IOException attempting to get > file from file:/var/spool/opencast/cache/schedule.ics. > 2012-01-06 08:36:39 WARN (IngestRestService:506) - workflowInstanceId > 'Unscheduled-MH-CA-01-1325860522721' is not numeric > 2012-01-06 08:35:07 WARN (CaptureAgentStateServiceImpl:421) - Recording id > 'Unscheduled-MH-CA-01-1325860522721' is not a long, and is therefore not a > valid workflow identifier > 2012-01-06 08:32:40 ERROR (?:?) - > [org.opencastproject.scheduler.impl.SchedulerServiceImpl] Component instance > could not be created, activation failed > 2012-01-06 08:32:40 ERROR (?:?) - > [org.opencastproject.scheduler.impl.SchedulerServiceImpl] The activate method > has thrown an exception > org.osgi.framework.ServiceException: Exception [EclipseLink-4002] (Eclipse > Persistence Services - 2.0.2.v20100323-r6872): > org.eclipse.persistence.exceptions.DatabaseException > Internal Exception: > com.mysql.jdbc.exceptions.jdbc4.MySQLSyntaxErrorException: Unknown column > 'CA_METADATA' in 'field list' > Error Code: 1054 > Call: SELECT EVENT_ID, CA_METADATA, DUBLIN_CORE FROM SCHED_EVENT > Query: ReadAllQuery(referenceClass=EventEntity sql="SELECT EVENT_ID, > CA_METADATA, DUBLIN_CORE FROM SCHED_EVENT") > at > org.opencastproject.scheduler.impl.SchedulerServiceImpl.activate(SchedulerServiceImpl.java:190) > ... > at > org.apache.felix.fileinstall.internal.DirectoryWatcher.run(DirectoryWatcher.java:241) > 2012-01-06 08:32:40 WARN (SchedulerServiceImpl:189) - Unable to index event > instances: Exception [EclipseLink-4002] (Eclipse Persistence Services - > 2.0.2.v20100323-r6872): org.eclipse.persistence.exceptions.DatabaseException > Internal Exception: > com.mysql.jdbc.exceptions.jdbc4.MySQLSyntaxErrorException: Unknown column > 'CA_METADATA' in 'field list' > Error Code: 1054 > Call: SELECT EVENT_ID, CA_METADATA, DUBLIN_CORE FROM SCHED_EVENT > Query: ReadAllQuery(referenceClass=EventEntity sql="SELECT EVENT_ID, > CA_METADATA, DUBLIN_CORE FROM SCHED_EVENT") > 2012-01-06 08:32:40 ERROR (SchedulerServiceDatabaseImpl:218) - Could not > retrieve all events: > Internal Exception: > com.mysql.jdbc.exceptions.jdbc4.MySQLSyntaxErrorException: Unknown column > 'CA_METADATA' in 'field list' > Error Code: 1054 > Call: SELECT EVENT_ID, CA_METADATA, DUBLIN_CORE FROM SCHED_EVENT > Query: ReadAllQuery(referenceClass=EventEntity sql="SELECT EVENT_ID, > CA_METADATA, DUBLIN_CORE FROM SCHED_EVENT") > 2012-01-06 08:32:39 INFO (WorkflowDefinitionScanner:55) - Installing > workflow from file > /opt/matterhorn/felix/conf/workflows/default-error-handler.xml > 2012-01-06 08:32:37 WARN (TrustedHttpClientImpl:246) - URI > http://monssilvius.lsu.edu:8080/capture-admin/agents/demo_capture_agent/configuration > does not support digest authentication > 2012-01-06 08:32:37 WARN (TrustedHttpClientImpl:246) - URI > http://monssilvius.lsu.edu:8080/capture-admin/agents/demo_capture_agent does > not support digest authentication > 2012-01-06 08:32:33 WARN (ConfigurationManager:126) - Malformed URL for > capture.config.remote.endpoint.url, disabling polling. > > > CA: > ============ > 2012-01-06 08:44:49 WARN (SchedulerImpl:599) - Invalid calendar data, > skipping parse attempt. > 2012-01-06 08:44:49 WARN (IoSupport:241) - IOException attempting to get > file from file:/var/spool/opencast/schedule.ics. > 2012-01-06 08:39:49 WARN (SchedulerImpl:599) - Invalid calendar data, > skipping parse attempt. > 2012-01-06 08:39:49 WARN (IoSupport:241) - IOException attempting to get > file from file:/var/spool/opencast/schedule.ics. > 2012-01-06 08:36:03 WARN (CaptureAgentImpl:918) - Confidence monitoring not > started: {} > java.lang.NullPointerException > at > org.opencastproject.capture.impl.CaptureAgentImpl.setAgentState(CaptureAgentImpl.java:905) > at > org.opencastproject.capture.impl.CaptureAgentImpl.stopCapture(CaptureAgentImpl.java:506) > at > org.opencastproject.capture.endpoint.CaptureRestService.stopCapture(CaptureRestService.java:176) > ... > at > org.mortbay.thread.QueuedThreadPool$PoolThread.run(QueuedThreadPool.java:582) > 2012-01-06 08:34:48 WARN (CaptureAgentImpl:918) - Confidence monitoring not > started: {} > java.lang.NullPointerException > at > org.opencastproject.capture.impl.CaptureAgentImpl.setAgentState(CaptureAgentImpl.java:905) > at > org.opencastproject.capture.impl.CaptureAgentImpl.activate(CaptureAgentImpl.java:1249) > ... > at > org.apache.felix.fileinstall.internal.DirectoryWatcher.run(DirectoryWatcher.java:241) > 2012-01-06 08:34:48 WARN (ConfigurationManager:126) - Malformed URL for > capture.config.remote.endpoint.url, disabling polling. > Jan 6 08:34:23 MH-CA-01 NetworkManager[774]: <error> [1325860463.335305] > [nm-manager.c:1317] user_proxy_init(): could not init user settings proxy: > (3) Could not get owner of name 'org.freedesktop.NetworkManagerUserSettings': > no such name > Jan 6 08:34:23 MH-CA-01 NetworkManager[774]: <error> [1325860463.290390] > [nm-manager.c:1317] user_proxy_init(): could not init user settings proxy: > (3) Could not get owner of name 'org.freedesktop.NetworkManagerUserSettings': > no such name > Jan 6 08:34:23 MH-CA-01 NetworkManager[774]: <error> [1325860463.335305] > [nm-manager.c:1317] user_proxy_init(): could not init user settings proxy: > (3) Could not get owner of name 'org.freedesktop.NetworkManagerUserSettings': > no such name > Jan 6 08:34:23 MH-CA-01 NetworkManager[774]: <error> [1325860463.290390] > [nm-manager.c:1317] user_proxy_init(): could not init user settings proxy: > (3) Could not get owner of name 'org.freedesktop.NetworkManagerUserSettings': > no such name > On Jan 5, 2012, at 3:58 PM, Greg Logan wrote: > >> On 12-01-05 02:32 PM, Jason R Peak wrote: >>> Thanks for having a look Greg. >>> Your questions may have uncovered some deeper issues: >>> >>>> Do the tables >>>> get created in the DB? >>> I used the DDL script, and sometime early on, I flushed some of the tables >>> of old failed jobs. >>> Can I verify anything about the DB? >> >> Hrm, I'm not 100% sure that the DDL scripts work because I haven't >> worked with 1.2 for a while! Can you try it using the autogenerated >> tables? Wipe the DB, and turn on org.opencastproject.db.ddl.generation >> in config.properties. >> >>>> Are you sure that the >>>> matterhorn process(es) can write to /var/spool/opencast*? >>> on both machines, the /var/spool/opencast directory is owned, user and >>> group, by the matterhorn user with recursive permissions 755. >>> Additionally, neither machine has any trouble writing to the subdirectories >>> of /var/spool/opencast. >>> Should there be a scheduler service endpoint somewhere? I have many >>> services listed on my core's welcome page? >>> AHA! on the opencast demo, there IS a scheduler service listed. >>> Furthermore, the following URL returns 404 for my core: >>> /scheduler/all/events.xml >>> How can I reconnect my scheduler? >>> >>>> Do you see any ERRORs or other odd >>>> messages in the logs of either machine? >>> >>> Greg, for the CA, this is quite a long listing, so I have elided the >>> middles of these 97-line-long stack traces (will provide the source, if >>> useful): >> >> It looks to me like some of the bundles aren't starting correctly on >> both your core and your CA, so let's try rebuilding them and making sure >> that the build finishes successfully. Stop Matterhorn, then in the >> checked out source code directory run the following before starting >> Matterhorn again: >> >> CA: mvn clean install -Pcapture,serviceregistry-stub >> -DdeployTo=$FELIX_HOME/matterhorn >> Core: mvn clean install -DdeployTo=$FELIX_HOME/matterhorn >> >> If that doesn't work then please stop Matterhorn, delete >> $FELIX_HOME/felix-cache, and then restart Matterhorn. >> >> G >> >>> for the CA: >>> ********************************* >>> 2012-01-05 12:41:33 WARN (CaptureAgentImpl:918) - Confidence monitoring >>> not started: {} >>> java.lang.NullPointerException >>> at >>> org.opencastproject.capture.impl.CaptureAgentImpl.setAgentState(CaptureAgentImpl.java:905) >>> at >>> org.opencastproject.capture.impl.CaptureAgentImpl.stopCapture(CaptureAgentImpl.java:506) >>> at >>> org.opencastproject.capture.endpoint.CaptureRestService.stopCapture(CaptureRestService.java:176) >>> ... >>> at >>> org.mortbay.thread.QueuedThreadPool$PoolThread.run(QueuedThreadPool.java:582) >>> 2012-01-05 12:21:14 WARN (CaptureAgentImpl:918) - Confidence monitoring >>> not started: {} >>> java.lang.NullPointerException >>> at >>> org.opencastproject.capture.impl.CaptureAgentImpl.setAgentState(CaptureAgentImpl.java:905) >>> at >>> org.opencastproject.capture.impl.CaptureAgentImpl.activate(CaptureAgentImpl.java:1249) >>> ... >>> at >>> org.apache.felix.fileinstall.internal.DirectoryWatcher.run(DirectoryWatcher.java:241) >>> 2012-01-05 12:21:14 WARN (ConfigurationManager:126) - Malformed URL for >>> capture.config.remote.endpoint.url, disabling polling. >>> 2012-01-05 12:20:54 WARN (ConfigurationManager:126) - Malformed URL for >>> capture.config.remote.endpoint.url, disabling polling. >>> 2012-01-05 12:03:05 WARN (CaptureAgentImpl:918) - Confidence monitoring >>> not started: {} >>> java.lang.NullPointerException >>> at >>> org.opencastproject.capture.impl.CaptureAgentImpl.setAgentState(CaptureAgentImpl.java:905) >>> at >>> org.opencastproject.capture.impl.CaptureAgentImpl.stopCapture(CaptureAgentImpl.java:506) >>> at >>> org.opencastproject.capture.endpoint.CaptureRestService.stopCapture(CaptureRestService.java:176) >>> ... >>> at >>> org.mortbay.thread.QueuedThreadPool$PoolThread.run(QueuedThreadPool.java:582) >>> 2012-01-05 11:31:40 WARN (CaptureAgentImpl:918) - Confidence monitoring >>> not started: {} >>> java.lang.NullPointerException >>> at >>> org.opencastproject.capture.impl.CaptureAgentImpl.setAgentState(CaptureAgentImpl.java:905) >>> at >>> org.opencastproject.capture.impl.CaptureAgentImpl.stopCapture(CaptureAgentImpl.java:506) >>> at >>> org.opencastproject.capture.endpoint.CaptureRestService.stopCapture(CaptureRestService.java:176) >>> ... >>> at >>> org.mortbay.thread.QueuedThreadPool$PoolThread.run(QueuedThreadPool.java:582) >>> >>> =================================== >>> for the core: >>> *********************** >>> 2012-01-05 12:41:42 WARN (IngestRestService:506) - workflowInstanceId >>> 'Unscheduled-MH-CA-01-1325788871971' is not numeric >>> 2012-01-05 12:40:57 WARN (CaptureAgentStateServiceImpl:421) - Recording id >>> 'Unscheduled-MH-CA-01-1325788871971' is not a long, and is therefore not a >>> valid workflow identifier >>> 2012-01-05 12:38:41 WARN (SeriesRestService:524) - No sort enum matches >>> 'TITLE_ASC' >>> 2012-01-05 12:03:41 WARN (IngestRestService:506) - workflowInstanceId >>> 'Unscheduled-MH-CA-01-1325786565076' is not numeric >>> 2012-01-05 12:02:29 WARN (CaptureAgentStateServiceImpl:421) - Recording id >>> 'Unscheduled-MH-CA-01-1325786565076' is not a long, and is therefore not a >>> valid workflow identifier >>> 2012-01-05 11:42:54 WARN (SeriesRestService:524) - No sort enum matches >>> 'TITLE_ASC' >>> 2012-01-05 11:31:41 WARN (IngestRestService:506) - workflowInstanceId >>> 'Unscheduled-MH-CA-01-1325784691259' is not numeric >>> 2012-01-05 11:31:19 WARN (CaptureAgentStateServiceImpl:421) - Recording id >>> 'Unscheduled-MH-CA-01-1325784691259' is not a long, and is therefore not a >>> valid workflow identifier >>> >>> =================================== >>> >>> >>> Thanks Greg! >>> Jason >>> >>> On Jan 5, 2012, at 1:57 PM, Greg Logan wrote: >>> >>>> On 12-01-05 01:43 PM, Jason R Peak wrote: >>>>> Sure thing Greg, thanks. >>>>> Default creds on the CA, although I HAVE gotten CAS working for the >>>>> core... >>>>> >>>>> I have certainly not clearly understood the capture agent config. >>>>> Config for both hosts are attached; I have NOT modified >>>>> CORE...ConfigurationManager.properties. >>>>> Jason >>>> >>>> All of those config files look correct to me, and they're certainly free >>>> from the usual newbie mistakes. Do you see any ERRORs or other odd >>>> messages in the logs of either machine? Are you sure that the >>>> matterhorn process(es) can write to /var/spool/opencast*? Do the tables >>>> get created in the DB? >>>> >>>> G >>>> >>>>> On 1/5/12 1:30 PM, "Greg Logan" <[email protected]> wrote: >>>>> >>>>>> On 12-01-05 01:23 PM, Jason R Peak wrote: >>>>>>> Hello All, >>>>>>> >>>>>>> >>>>>>> I am working towards a proof-of-concept installation of Matterhorn 1.2, >>>>>>> and I have been getting the following errors for quite awhile re: the >>>>>>> scheduling service. >>>>>>> >>>>>>> >>>>>>> 2012-01-05 12:48:47 WARN (IoSupport:241) - IOException attempting to >>>>>>> get file from file:/var/spool/opencast/cache/schedule.ics. >>>>>>> >>>>>>> 2012-01-05 12:48:47 WARN (SchedulerImpl:599) - Invalid calendar data, >>>>>>> skipping parse attempt. >>>>>>> >>>>>>> 2012-01-05 12:53:47 WARN (IoSupport:241) - IOException attempting to >>>>>>> get file from file:/var/spool/opencast/cache/schedule.ics. >>>>>>> >>>>>>> 2012-01-05 12:53:47 WARN (SchedulerImpl:599) - Invalid calendar data, >>>>>>> skipping parse attempt. >>>>>>> >>>>>>> 2012-01-05 12:58:47 WARN (IoSupport:241) - IOException attempting to >>>>>>> get file from file:/var/spool/opencast/cache/schedule.ics. >>>>>>> >>>>>>> 2012-01-05 12:58:47 WARN (SchedulerImpl:599) - Invalid calendar data, >>>>>>> skipping parse attempt. >>>>>>> >>>>>>> 2012-01-05 13:03:47 WARN (IoSupport:241) - IOException attempting to >>>>>>> get file from file:/var/spool/opencast/cache/schedule.ics. >>>>>>> >>>>>>> 2012-01-05 13:03:47 WARN (SchedulerImpl:599) - Invalid calendar data, >>>>>>> skipping parse attempt. >>>>>>> >>>>>>> >>>>>>> I have never been able to schedule a recording successfully; I can >>>>>>> submit the form, but nothing shows up in the 'Upcoming' tab of the >>>>>>> recordings listing (I am assuming that it would be thereŠ). >>>>>> >>>>>> From this description it sounds like both the core and CA are damaged: >>>>>> It should be able to schedule even if the capture doesn't fire. Can you >>>>>> please attach your config.properties and >>>>>> org.opencastproject.capture.impl.ConfigurationManager.properties files >>>>>> from both your core and CA? Don't forget to blank out any passwords! >>>>>> >>>>>> G >>>>>> >>>>>>> These messages show up in the core AND in the Capture agent, suggesting >>>>>>> to me that my CA config is off. >>>>>>> >>>>>>> Has anyone had similar Issues? >>>>>>> >>>>>>> Thanks, >>>>>>> Jason >>>>>>> >>>>>>> >>>>>>> _______________________________________________ >>>>>>> Matterhorn-users mailing list >>>>>>> [email protected] >>>>>>> http://lists.opencastproject.org/mailman/listinfo/matterhorn-users >>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> Matterhorn-users mailing list >>>>>> [email protected] >>>>>> http://lists.opencastproject.org/mailman/listinfo/matterhorn-users >>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> Matterhorn-users mailing list >>>>>> [email protected] >>>>>> http://lists.opencastproject.org/mailman/listinfo/matterhorn-users >>>> >>>> >>>> _______________________________________________ >>>> Matterhorn-users mailing list >>>> [email protected] >>>> http://lists.opencastproject.org/mailman/listinfo/matterhorn-users >>> >>> >>> _______________________________________________ >>> Matterhorn-users mailing list >>> [email protected] >>> http://lists.opencastproject.org/mailman/listinfo/matterhorn-users >> >> >> _______________________________________________ >> Matterhorn-users mailing list >> [email protected] >> http://lists.opencastproject.org/mailman/listinfo/matterhorn-users > > > _______________________________________________ > Matterhorn-users mailing list > [email protected] > http://lists.opencastproject.org/mailman/listinfo/matterhorn-users
signature.asc
Description: OpenPGP digital signature
_______________________________________________ Matterhorn-users mailing list [email protected] http://lists.opencastproject.org/mailman/listinfo/matterhorn-users
