Hello, We stumbled across what looks like a pretty serious issue with 7.x and 8.x versions of Solr. There is a hardcoding of certain timeouts in the code due to which replica recoveries in Solr do not happen at all. This issue is mostly limited to heavily loaded clusters where *:* query of the ping handler takes several seconds to complete. Jira for the same is https://issues.apache.org/jira/browse/SOLR-13532 I would like to hear what other devs think of this.
Thanks, Suril
