# \_refresh taking 30+ seconds **URL:** \ **Category:** Elasticsearch\ **Created:** [October 26, 2016, 4:09pm UTC](https://discuss.elastic.co/t/-refresh-taking-30-seconds/64028 "2016-10-26T16:09:13Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![apecoraro](https://avatars.discourse-cdn.com/v4/letter/a/ac91a4/32.png) [@apecoraro](https://discuss.elastic.co/u/apecoraro)\ **Post date:** [October 26, 2016, 4:09pm UTC](https://discuss.elastic.co/t/-refresh-taking-30-seconds/64028/1 "2016-10-26T16:09:14Z")
Is it normal for a POST \_refresh on a single index to take longer than 30 seconds? That's what I see happening consistently on my 5 node ES cluster (running 5.0.0-rc1). My application needs to call refresh because it needs to be able to search on documents that it has just inserted. Since the ES refresh rate is supposed to be 1 second. Would it be safe for me to just have my app sleep for a few seconds (rather than waiting 30 seconds for the POST \_refresh to come back). It also only seems to happen on one particular index. The other two indexes \_refresh very quickly. And the slow index also only has a single document in it. Plus, I just discovered that index, update, and delete operations using refresh=true take a fraction of second to complete. So I guess that's a workable work around, but I'd still like to understand why it would take so long to do POST \_refresh on that index. Thanks. ---
**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, 10:09pm UTC](https://discuss.elastic.co/t/-refresh-taking-30-seconds/64028/2 "2017-07-05T22:09:22Z")