There may be improvements in this area since Solr 8.11, but last time I'd seen it, I worked-around it by switching to http1. See this thread: https://www.mail-archive.com/users@solr.apache.org/msg03272.html
On Thu, Jun 27, 2024 at 10:25 AM Paul Cote <paul.c...@hitachivantara.com.invalid> wrote: > To answer my own inquiry. This is a solr resource (jvm) depletion issue. > Restarting each node in the cluster "Solved" it. > Ref: > https://www.searchstax.com/docs/hc/max-requests-queued-per-destination-3000-exceeded/ > > On 2024/06/25 15:25:19 Paul Cote wrote: > > Hello > > > > Curious problem just surfaced after several months of using solr V8.11.1 > > > > I'm seeing these messages, repeatedly, when accessing indexes on just 2 > of 6 nodes running solr. All nodes are identical wrt to O/S, docker, solr, > ZK, etc > > > > > 4 of the nodes functioning just fine, just 2 are behaving this way, > regardless of which shards are local to those nodes. Sockets are not hung. > SAR network stats shows all is fine. > > > > > Any insight / guidance would be greatly appreciated. Thanks > > > > org.apache.solr.common.SolrException: > org.apache.solr.client.solrj.SolrServerException: Max requests queued per > destination 3000 exceeded for HttpDestination[http://w.x.y.z/etc > > > > > > > >
