Trying to install and configure elastic enterprise search connector(self managed) to connect with share point online.
Using source code to run the connector client.
While setup can be able to create the executable files for the connector client, but while start the connector(./bin/connector) getting the below exception Exception:
Traceback (most recent call last):
File "./bin/connectors", line 33, in <module>
sys.exit(load_entry_point('elasticsearch-connectors', 'console_scripts', 'connectors')())
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/./bin/connectors", line 22, in importlib_load_entry_point
for entry_point in distribution(dist_name).entry_points
^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/lib64/python3.12/importlib/metadata/__init__.py", line 861, in distribution
return Distribution.from_name(distribution_name)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/lib64/python3.12/importlib/metadata/__init__.py", line 399, in from_name
raise PackageNotFoundError(name)
importlib.metadata.PackageNotFoundError: No package metadata was found for elasticsearch-connectors
Sorry you're having issues running the connectors from source.
Have you tried running make clean install since hitting this issue? This looks like the type of error that we sometimes see when we change branches or make code changes but forget to re-install packages.
Make sure that make clean install completes without errors.
If you're still having the issue after that, can you tell me:
Thanks for your quick response.
With the make clean install, it is working fine.
we can be able to create connector and job for the service type sharepoint_online.
If we view the Job from the Job id, the status showing is pending, no documents getting indexed but index is created.
Can you please help us on how to debug more or how to check the logs for the same.
Note: We are using Elastic trial version.
Is this due to any licensing issue?
command used to create Job - ./bin/connectors job start -i GAhKZo4B8S1noEQReDNs -t full -o json
Shouldn't be, no. The only things tied to a paid license would be Document Level Security for the SPO connector, but even then I don't believe it would just hang for you.
the status showing is pending, no documents getting indexed but index is created.
To diagnose this, you'll want to look at the connector framework's logs. You may also want to consider changing the log level to DEBUG if you don't initially have enough information.
We did enable the DEBUG logging in config.yml(service.log_level:DEBUG) inside the connectors directory. However, we do not see any logs being generated.
We disabled Enumerate all sites, Subsites but still no luck.
How are you running connectors? Are you running from the source code with make run or are you using docker images?
Do you not see any logs, or just didn't see DEBUG logs after making the change?
Did you restart the connectors process?
The connector process produces logs. So we just need to validate that you're actually starting it. If it's not running, that would explain why your jobs always stay pending
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.