Closing stale connection with Openlink Virtuoso Docker image version > 7.2.7

[mhoangvslev]

While using rdf4j, I encountered this error, which I reported on their repo.

It only happens on Virtuoso 7.2.8 and above.

I came across this forum :

At the HTTP level a connection can often be reused to reduce latency, but sometimes this connection gets closed by the server while the client still thinks the connection is open.

Older versions of Apache HttpClient checked for stale connections themselves, and simply opened a new one when necessary. For the current version we had to add a custom retry handler that serves the same purpose, and this handler logs these warnings.

Is there anything I can do from virtuoso.ini file to workaround this?

[mhoangvslev]

I compared the virtuoso.ini of 7.2.7 and 7.2.9, and the only difference is the GraphQL plugin (issue #1109)

[HughWilliams]

There are no settings in the INI file to control such behaviour to my knowledge.

What is the RDF4J app being used is it the RDF4J FedX service referenced in git issue https://github.com/openlink/virtuoso-opensource/issues/1109) ?

Note Virtuoso has its own RDF4J Provider and we have sample RDF4J programs available with best usage examples.

[mhoangvslev]

I follow their instructions here to create an app