Hi, We are seeing transient Connection reset in our custom solr client(a wrapper around solrj). We want to add retries to all methods that we are currently using so that the we are able to upload successfully. However, I'm not sure if there's any relevant documentation on which methods are idempotent and which aren't.
Our use case - We have a single solr host. We aren't using solr cloud or anything fancy, We want to upload an index to Solr host. To do that, we first: 1) Disable replication 2) delete old index 3) upload new index 4) commit the changes (rollback if there's an exception) 5) run a solr query and perform some validations 6) run /admin/luke and perform some validation. 7) enable replication We're currently thinking it should be OK to retry each of these 6 requests(atleast for Socket Exceptions), but would like guidance/confirmation. Any documentation on this would be really helpful. Thanks -- Pranay
