[SOLR-13474] Fix "Search is temporarily disabled" logic to be consistent for entire request

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 8.2, 9.0
• Assignee: Chris M. Hostetter
• Reporter: Chris M. Hostetter
• Created: 16/May/19 05:45
• Updated: 26/Jul/19 08:56
• Resolved: 20/May/19 18:00

Description

Where/How the "Search is temporarily disabled" logic was added in SOLR-12999 can result in some weird failure cases (examples: SOLR-13470 & SOLR-13471) due to the fact that  (as currently implemented) some portions of processing a SolrQueryRequest may see a valid searcher, but later other code paths will get an error when trying to use the same searcher.

I think it would make more sense to move the "searchEnabled" logic out of SolrQueryRequestbase.getSearcher() and into SolrCore.getSearcher() (removing the need for SolrCore.isSearchEnabled() ).

This will:

• restore the long standing (and widly assumed by code) behavior of SolrQueryRequestbase.getSearcher() that it will always return a consistent result
• ensure that any custom plugins trying to call SolrCore.getSearcher() directly will also get the benefits of the safety checks introduced in SOLR-12999

Issue Links

• causes: SOLR-13470 SolrException msg not always propogated to HttpClient if exception occurs during response writting
• incorporates: SOLR-13469 rejected requests during ful IndexFetch should not use 403 response code
• relates to: SOLR-12999 Index replication could delete segments first