# SAN index replication from PROD to DR(disaster recovery) server **URL:** \ **Category:** Elasticsearch\ **Created:** [April 20, 2016, 10:14am UTC](https://discuss.elastic.co/t/san-index-replication-from-prod-to-dr-disaster-recovery-server/47894 "2016-04-20T10:14:14Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![satyapinku](https://avatars.discourse-cdn.com/v4/letter/s/3e96dc/32.png) [@satyapinku](https://discuss.elastic.co/u/satyapinku)\ **Post date:** [April 20, 2016, 10:14am UTC](https://discuss.elastic.co/t/san-index-replication-from-prod-to-dr-disaster-recovery-server/47894/1 "2016-04-20T10:14:14Z")
Hi, I am facing few issues in replicating my SAN index to DR server from my PROD index. For NAS index its perfectly fine we are able to replicate but for SAN we are not. We have used the OS rsync method to copy index from PROD to DR but it fails sometimes. Is their any way to do SAN replication. Any guidance and help is welcome P.S. PROD and DR are different servers Thanks Satya ---
**Author:** ![Bruce\_Ritchie](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/bruce_ritchie/32/9370_2.png) [@Bruce\_Ritchie](https://discuss.elastic.co/u/Bruce_Ritchie)\ **Post date:** [April 20, 2016, 9:29pm UTC](https://discuss.elastic.co/t/san-index-replication-from-prod-to-dr-disaster-recovery-server/47894/2 "2016-04-20T21:29:06Z")
Perhaps periodic snapshots from ES may be a better approach - [https://www.elastic.co/guide/en/elasticsearch/guide/current/backing-up-your-cluster.html](https://www.elastic.co/guide/en/elasticsearch/guide/current/backing-up-your-cluster.html) [https://www.elastic.co/guide/en/elasticsearch/reference/current/modules-snapshots.html](https://www.elastic.co/guide/en/elasticsearch/reference/current/modules-snapshots.html) Using rsync I think would generally be problematic unless you can flush and create an filesystem-level snapshot (ala ZFS) and rsync that over to the DR server. Otherwise, you're attempting to sync files that are in a constant state of flux. ---
**Author:** ![satyapinku](https://avatars.discourse-cdn.com/v4/letter/s/3e96dc/32.png) [@satyapinku](https://discuss.elastic.co/u/satyapinku)\ **Post date:** [April 21, 2016, 2:45pm UTC](https://discuss.elastic.co/t/san-index-replication-from-prod-to-dr-disaster-recovery-server/47894/3 "2016-04-21T14:45:32Z")
Hi Bruce, Snapshots is fine but we have index in SAN(local drive) and not NAS(network drive). So how will we replicate it from PROD to DR as SAN is not over network. Do you mean to say create local snapshots in PROD and then manually push them to DR everyday ? Thanks Satya ---
**Author:** ![Bruce\_Ritchie](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/bruce_ritchie/32/9370_2.png) [@Bruce\_Ritchie](https://discuss.elastic.co/u/Bruce_Ritchie)\ **Post date:** [April 21, 2016, 2:55pm UTC](https://discuss.elastic.co/t/san-index-replication-from-prod-to-dr-disaster-recovery-server/47894/4 "2016-04-21T14:55:38Z")
Satya, That was my first thought, yes. Have a local snapshot directory in PROD that you rsync to DR and use that for restoring the index into DR. It may not be the most efficient method but I'm fairly certain it'll be consistent and stable. If you need more real-time DR than what you can get with snapshot/restore you may have to introduce middleware to synchronize the ES instances. See [https://www.elastic.co/blog/scaling\_elasticsearch\_across\_data\_centers\_with\_kafka](https://www.elastic.co/blog/scaling_elasticsearch_across_data_centers_with_kafka) for an example of what that might look like. Bruce ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [July 5, 2017, 10:57pm UTC](https://discuss.elastic.co/t/san-index-replication-from-prod-to-dr-disaster-recovery-server/47894/5 "2017-07-05T22:57:16Z")