Uploaded image for project: 'Nuxeo Platform'
  1. Nuxeo Platform
  2. NXP-17411

Repository Elasticsearch re-indexing drops audit log entries

    XMLWordPrintable

    Details

      Description

      When using audit on Elasticsearch backend reindexing from the admin center will drop audit and sequence index:

      2015-06-26 16:51:37,796 WARN  [http-bio-0.0.0.0-8080-exec-5] [org.nuxeo.elasticsearch.web.admin.ElasticSearchManager] Re-indexing the entire repository: default
      2015-06-26 16:51:37,805 WARN  [http-bio-0.0.0.0-8080-exec-5] [org.nuxeo.elasticsearch.core.ElasticSearchAdminImpl] Initializing index: sequence, type: seqId with dropIfExists flag, deleting an existing index
      2015-06-26 16:51:37,932 WARN  [http-bio-0.0.0.0-8080-exec-5] [org.nuxeo.elasticsearch.core.ElasticSearchAdminImpl] Initializing index: audit, type: entry with dropIfExists flag, deleting an existing index
      2015-06-26 16:51:38,579 WARN  [http-bio-0.0.0.0-8080-exec-5] [org.nuxeo.elasticsearch.core.ElasticSearchAdminImpl] Initializing index: nuxeo, type: doc with dropIfExists flag, deleting an existing index
      

      This is a problem as since 7.3 Elasticsearch is used as a default backend for audit logs, see NXP-17236, so the audit index becomes the reference and should never be dropped.
      neither the related uidgen index.

      Waiting for the issue to be fixed, two possible workarounds:

      1. Use the attached patched jar.
      2. Backup the audit Elasticsearch index before performing an entire re-indexation and restore it afterwards.

        Attachments

          Issue Links

            Activity

              People

              • Votes:
                0 Vote for this issue
                Watchers:
                4 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved: