Re: [OC Users] Text extraction problem

[Greg Logan]

Hi Hartmut,

The URLs in the log should work, do they?  If not then there's likely something wrong config wise.  Ideally the initial lookup via filesystem would work - it's going to be significantly faster.  Check that the NFS mount(s) are working correctly on the affected worker(s).

G

On Thu, Mar 16, 2023 at 6:42 AM Hartmut Trüe <htr...@gmail.com> wrote:

Hi,

i see many errors like these during processing on my worker node. Is it a misconfiguration?

2023-03-16T12:54:37,956 | INFO  | (TextAnalyzerServiceImpl:188) - Starting text extraction from https:// ADMIN.NODE/files/collection/composer/1379449_31.png
2023-03-16T12:54:37,957 | WARN  | (WorkspaceImpl:386) - The working file repository and workspace paths don't match. Looking up https://ADMIN.NODE/files/collection/composer/1379449_31.png at /srv/opencast/files/collection/composer/1379449_31.png failed
2023-03-16T12:54:37,963 | ERROR | (AbstractJobProducer$JobRunner:343) - Error handling operation 'Extract':
org.opencastproject.serviceregistry.api.ServiceRegistryException: Error handling operation 'Extract'
        at org.opencastproject.textanalyzer.impl.TextAnalyzerServiceImpl.process(TextAnalyzerServiceImpl.java:277) ~[?:?]
        at org.opencastproject.job.api.AbstractJobProducer$JobRunner.call(AbstractJobProducer.java:313) [!/:?]
        at org.opencastproject.job.api.AbstractJobProducer$JobRunner.call(AbstractJobProducer.java:272) [!/:?]
        at java.util.concurrent.FutureTask.run(FutureTask.java:264) [?:?]
        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1128) [?:?]
        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:628) [?:?]
        at java.lang.Thread.run(Thread.java:829) [?:?]
Caused by: org.opencastproject.textanalyzer.api.TextAnalyzerException: Image https://oc-admin.uni-hildesheim.de/files/collection/composer/1379449_9.png not found in workspace
        at org.opencastproject.textanalyzer.impl.TextAnalyzerServiceImpl.extract(TextAnalyzerServiceImpl.java:192) ~[?:?]
        at org.opencastproject.textanalyzer.impl.TextAnalyzerServiceImpl.process(TextAnalyzerServiceImpl.java:267) ~[?:?]
        ... 6 more
Caused by: org.opencastproject.util.NotFoundException
        at org.opencastproject.workspace.impl.WorkspaceImpl.downloadIfNecessary(WorkspaceImpl.java:534) ~[?:?]
        at org.opencastproject.workspace.impl.WorkspaceImpl$2.xapply(WorkspaceImpl.java:556) ~[?:?]
        at org.opencastproject.workspace.impl.WorkspaceImpl$2.xapply(WorkspaceImpl.java:553) ~[?:?]
        at org.opencastproject.util.data.Function$X.apply(Function.java:104) ~[!/:?]
        at org.opencastproject.util.IoSupport.locked(IoSupport.java:478) ~[!/:?]
        at org.opencastproject.workspace.impl.WorkspaceImpl.get(WorkspaceImpl.java:401) ~[?:?]
        at org.opencastproject.workspace.impl.WorkspaceImpl.get(WorkspaceImpl.java:354) ~[?:?]
        at org.opencastproject.textanalyzer.impl.TextAnalyzerServiceImpl.extract(TextAnalyzerServiceImpl.java:190) ~[?:?]
        at org.opencastproject.textanalyzer.impl.TextAnalyzerServiceImpl.process(TextAnalyzerServiceImpl.java:267) ~[?:?]
        ... 6 more

Regards,

Hartmut

--
To unsubscribe from this group and stop receiving emails from it, send an email to users+un...@opencast.org.

[Hartmut Trüe]

Hi Greg,

yes, it seems to be related to the distributed system with admin and worker node. It works fine on my single node test system.

The setting in org.opencastproject.organization-mh_default_org.cfg on all nodes is:

prop.org.opencastproject.file.repo.url= https://ADMIN.NODE

and the same property in custom.properties is commented out.

Should I change it on all nodes to

org.opencastproject.file.repo.path=${org.opencastproject.storage.dir}/files

which is the NFS mount /srv/opencast/files on all nodes?

Hartmut