# Elasticsearch process dead and http request stuck **URL:** \ **Category:** Elasticsearch\ **Created:** [December 30, 2017, 4:13am UTC](https://discuss.elastic.co/t/elasticsearch-process-dead-and-http-request-stuck/113622 "2017-12-30T04:13:15Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![zhaosi](https://avatars.discourse-cdn.com/v4/letter/z/b782af/32.png) [@zhaosi](https://discuss.elastic.co/u/zhaosi)\ **Post date:** [December 30, 2017, 4:13am UTC](https://discuss.elastic.co/t/elasticsearch-process-dead-and-http-request-stuck/113622/1 "2017-12-30T04:13:15Z")
I am running a 10 nodes es cluster, and everything is ok. yesterday I add 8 new nodes to the cluster, the new nodes work well at beginning, but after several hours later, all the new nodes disconnected from the cluster, and the http request to new nodes will get stuck, while the old nodes still work well. The es process in new nodes still exist but not ouput any log(include gc log). cannot be kill by sign 15. But when I use strace or jstack operate on es process, it back to noraml again, I tried this 4 times, and 4 times the process back to normal!!! info about os, jvm, es: os:Linux 2.6.32-504.el6.x86\_64 #1 SMP Wed Oct 15 04:27:16 UTC 2014 x86\_64 x86\_64 x86\_64 GNU/Linux jdk: 1.8.0\_131 es: 5.2.2 es config: ![](https://i.stack.imgur.com/eigYt.png) es jvm config: ![](https://i.stack.imgur.com/0vtAe.png) gc log: ![](https://i.stack.imgur.com/VtCMR.png) Spent whole day still cannot fix it, Anyone encounter this problem before? ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [January 27, 2018, 4:13am UTC](https://discuss.elastic.co/t/elasticsearch-process-dead-and-http-request-stuck/113622/2 "2018-01-27T04:13:16Z")
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.