sarkaramrit1993 opened a new pull request, #5330: URL: https://github.com/apache/solr/pull/5330
https://issues.apache.org/jira/browse/SOLR-18656 # Description When one replica's update stream stalls with updates still queued, `StreamingSolrClients.blockUntilFinished()` throws and `SolrCmdDistributor.finish()` only logs it. Retries are skipped and errors already collected from other replicas are dropped, so the leader returns success and failed replicas aren't put into a lower term. # Solution Catch the stall per client and report it through that client's `handleError`. The other clients are still waited on and retried. DBQ and commit used to fail the whole request on a stall; they now record an error for that node. A stream that stalls after its queue is drained isn't changed: stall detection needs a non-empty queue, so that still waits for the socket timeout. # Tests `SolrCmdDistributorTest.testStallKeepsErrorsFromOtherNodes`: one node never reads, another refuses. On main `getErrors()` is empty. Related: [SOLR-5939](https://issues.apache.org/jira/browse/SOLR-5939) / #5075 touches the same error path. -- 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]
