# Backup procedure for ES nodes
**URL:** https://discuss.elastic.co/t/backup-procedure-for-es-nodes/12616
**Category:** Elasticsearch
**Created:** [July 2, 2013, 1:51pm UTC](https://discuss.elastic.co/t/backup-procedure-for-es-nodes/12616 "2013-07-02T13:51:19Z")
**Posts on this page:** 3
**Page:** 1
### Author:  [@vijayprabhakar](https://discuss.elastic.co/u/vijayprabhakar)
#### Post date: [July 2, 2013, 1:51pm UTC](https://discuss.elastic.co/t/backup-procedure-for-es-nodes/12616/1 "2013-07-02T13:51:19Z")
Hi Team
I need some backup policies for my environment.
I am using 5 clusters with 3 nodes which is spread on 3 machines.
I need to take backup of all indexes in a cluster.
So what i thinking is shall i copy and archive the "data" directory alone..is this correct ..?
Also its enough to take the backup from any one node for a cluster, Whether using that i can able to recover the cluster ..?
---
### Author:  [@radu\_gheorghe](https://discuss.elastic.co/u/radu_gheorghe)
#### Post date: [July 10, 2013, 7:47am UTC](https://discuss.elastic.co/t/backup-procedure-for-es-nodes/12616/2 "2013-07-10T07:47:03Z")
Hello,
Before backing up your data directory, you need to disable flush, so you
don't catch it during a flush and get an inconsistent state:
curl -XPUT "$ADDRESS/$INDEX\_NAME/\_settings" -d '{
"index": {
"translog.disable\_flush": true
}}'
After your backup is done, you should enable flush back again.
You only need to do that one one node if that node contains a complete set
of shards. For example, if you have 3 nodes, and all your indices have 2
replicas, then it should work. If you have 0 or one replicas, it's best to
backup all data directories.
Here are two scripts I've tested for backing up and restoring "old" indices
containing logs, in case it gives any insight:
>
I was also optimizing them, because they wouldn't change from that point
on. I don't know it it makes sense for you or not.
Best regards,
Radu
On Tue, Jul 2, 2013 at 4:51 PM, vijayprabhakar
[vijayprabhakar.90@gmail.com](mailto:vijayprabhakar.90@gmail.com)wrote:
> Hi Team
>
> I need some backup policies for my environment.
> I am using 5 clusters with 3 nodes which is spread on 3 machines.
> I need to take backup of all indexes in a cluster.
>
> So what i thinking is shall i copy and archive the "data" directory
> alone..is this correct ..?
>
> Also its enough to take the backup from any one node for a cluster, Whether
> using that i can able to recover the cluster ..?
>
> --
> View this message in context:
> [http://elasticsearch-users.115913.n3.nabble.com/Backup-procedure-for-ES-nodes-tp4037374.html](http://elasticsearch-users.115913.n3.nabble.com/Backup-procedure-for-ES-nodes-tp4037374.html)
> Sent from the Elasticsearch Users mailing list archive at [Nabble.com](http://Nabble.com).
>
> --
> You received this message because you are subscribed to the Google Groups
> "elasticsearch" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com).
> For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out).
--
[http://sematext.com/](http://sematext.com/) -- Elasticsearch -- Solr -- Lucene
--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com).
For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out).
---
### Author:  [@system](https://discuss.elastic.co/u/system)
#### Post date: [July 6, 2017, 2:27am UTC](https://discuss.elastic.co/t/backup-procedure-for-es-nodes/12616/3 "2017-07-06T02:27:20Z")