janhoy opened a new issue, #116: URL: https://github.com/apache/solr-orbit/issues/116
`ClientFactory.create()` builds the data `SolrClient` without forwarding a `timeout`, so the client always uses the library's 30-second default no matter what `--client-options=timeout:N` says. The option parses, validates and is carried all the way to the factory, where it is dropped. Any single operation that legitimately takes longer than 30 seconds therefore cannot complete. The clearest case is `optimize` on a large index: a force-merge of a multi-hundred-gigabyte corpus runs for tens of minutes, the socket times out at 30 seconds, and the benchmark aborts with the merge still running server-side. Raising the timeout on the command line has no effect, which makes the failure hard to attribute — the operator has done exactly the documented thing. Bulk indexing is unaffected in practice only because each batch is individually fast; the exposure grows with corpus size and with any operation whose cost scales with the index. The factory should forward the configured timeout, falling back to a documented default when none is given, and warn rather than crash on an unparseable value. 🤖 Assisted by Claude Code -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
