# ES Master suddenly losing connection to data nodes

**URL:** <https://discuss.elastic.co/t/es-master-suddenly-losing-connection-to-data-nodes/25610>\
**Category:** Elasticsearch\
**Created:** [July 15, 2015, 10:18am UTC](https://discuss.elastic.co/t/es-master-suddenly-losing-connection-to-data-nodes/25610 "2015-07-15T10:18:51Z")\
**Posts on this page:** 1\
**Showing post:** 4

<div class="post-metadata">

**Author:** ![lwintergerst](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/lwintergerst/32/18164_2.png) [@lwintergerst](https://discuss.elastic.co/u/lwintergerst)\
**Post date:** [October 28, 2015, 3:05pm UTC](https://discuss.elastic.co/t/es-master-suddenly-losing-connection-to-data-nodes/25610/4 "2015-10-28T15:05:42Z")

</div>

hm.

you can check out my new post.

I think the solution is either the keepalive time of the tcp connections or a kernel bug.

> [@Random node disconnects - Java.io.IOException: Connection timed out](https://discuss.elastic.co/t/random-node-disconnects-java-io-ioexception-connection-timed-out/33151/1):
>
> EDIT 3: Solution TLDR smile : I found a way to finally solve this. This blog post describes it very well. My linux kernel has a bug with the network scatter/gather functionality. Turning this off solves this issue completely! EDIT 2: First try - this was not the Problem!! the OS was closing connections that were still needed. sysctl -w net.ipv4.tcp\_keepalive\_time=600 net.ipv4.tcp\_keepalive\_intvl=60 net.ipv4.tcp\_keepalive\_probes=3 found at We just recently started a ne…

---

_[View the full topic](https://discuss.elastic.co/t/es-master-suddenly-losing-connection-to-data-nodes/25610)._
