[SOLR-5890] Delete silently fails if not sent to shard where document was added

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 4.7
• Fix Version/s: 5.1, 6.0
• Component/s: SolrCloud
• Labels: difficulty-medium impact-medium workaround-exists

Environment

Debian 7.4.

• Assignee: Noble Paul Assign to me
• Reporter: Peter Inglesby
• Created: 20/Mar/14 15:39
• Updated: 02/Oct/19 17:24
• Resolved: 09/Feb/15 18:00

Description

We have SolrCloud set up with two shards, each with a leader and a replica. We use haproxy to distribute requests between the four nodes.

Regardless of which node we send an add request to, following a commit, the newly-added document is returned in a search, as expected.

However, we can only delete a document if the delete request is sent to a node in the shard where the document was added. If we send the delete request to a node in the other shard (and then send a commit) the document is not deleted. Such a delete request will get a 200 response, with the following body:

{'responseHeader'=>{'status'=>0,'QTime'=>7}}

Apart from the the very low QTime, this is indistinguishable from a successful delete.

Issue Links

• is duplicated by: SOLR-6984 Solr commitwithin is not happening for deletebyId
• is related to: SOLR-8889 SolrCloud deleteById is broken when router.field is set
• is related to: SOLR-12694 JavaBinUpdateRequestCodec fails to restore UpdateRequest.deleteById correctly when '_route_' is specified.
• is related to: SOLR-6910 deleteById without _route_ param for implicit router could be broadcast