# Very high sys cpu usage with HTTP KeepAlive **URL:** \ **Category:** Elasticsearch\ **Created:** [February 19, 2014, 12:38pm UTC](https://discuss.elastic.co/t/very-high-sys-cpu-usage-with-http-keepalive/15895 "2014-02-19T12:38:48Z")\ **Posts on this page:** 12\ **Page:** 1
**Author:** ![K\_Cima](https://avatars.discourse-cdn.com/v4/letter/k/58956e/32.png) [@K\_Cima](https://discuss.elastic.co/u/K_Cima)\ **Post date:** [February 19, 2014, 12:38pm UTC](https://discuss.elastic.co/t/very-high-sys-cpu-usage-with-http-keepalive/15895/1 "2014-02-19T12:38:48Z")
Hello, I am using ElasticSearch 1.0.0 with Kibana 3 on Solaris 11.1. My environment is as following. $ ./elasticsearch -v Version: 1.0.0, Build: a46900e/2014-02-12T16:18:34Z, JVM: 1.7.0\_07 $ java -version java version "1.7.0\_07" Java(TM) SE Runtime Environment (build 1.7.0\_07-b10) Java HotSpot(TM) Server VM (build 23.3-b01, mixed mode) $ cat /etc/release Oracle Solaris 11.1 X86 Copyright (c) 1983, 2012, Oracle and/or its affiliates. All rights reserved. Assembled 19 September 2012 I was suffering from very high CPU usage of Java process. When I connected to ES via Kibana, the CPU usage became 90%-100%. Strangely, user cpu usage was not so high, but sys cpu usage was very high with vmstat command. I tried to examine system calls by using truss command. I found there are so many ioctl calls while connecting. $ sudo truss -c -p 18496 syscall seconds calls errors read .000 13 4 write .000 12 ioctl 1.043 142731 lwp\_cond\_wait .002 178 153 lwp\_cond\_signal .000 25 pollsys .000 54 lwp\_mutex\_timedlock .000 1 pread64 .016 1432 -------- ------ ---- sys totals: 1.062 144446 157 usr time: .387 elapsed: 3.620 $ sudo truss -p 18496 ... /47: ioctl(99, DP\_POLL, 0xF6CB5750) = 1 /48: ioctl(102, DP\_POLL, 0xF6C746D0) = 1 /49: ioctl(105, DP\_POLL, 0xF6C33650) = 1 /48: ioctl(102, DP\_POLL, 0xF6C746D0) = 1 /47: ioctl(99, DP\_POLL, 0xF6CB5750) = 1 /48: ioctl(102, DP\_POLL, 0xF6C746D0) = 1 /47: ioctl(99, DP\_POLL, 0xF6CB5750) = 1 /49: ioctl(105, DP\_POLL, 0xF6C33650) = 1 /47: ioctl(99, DP\_POLL, 0xF6CB5750) = 1 /48: ioctl(102, DP\_POLL, 0xF6C746D0) = 1 /49: ioctl(105, DP\_POLL, 0xF6C33650) = 1 /47: ioctl(99, DP\_POLL, 0xF6CB5750) = 1 /48: ioctl(102, DP\_POLL, 0xF6C746D0) = 1 /49: ioctl(105, DP\_POLL, 0xF6C33650) = 1 /47: ioctl(99, DP\_POLL, 0xF6CB5750) = 1 /48: ioctl(102, DP\_POLL, 0xF6C746D0) = 1 /47: ioctl(99, DP\_POLL, 0xF6CB5750) = 1 ... I thought there are something wrong with HTTP KeepAlive. And, I considered to put HTTP Proxy between ES and Kibana(client browser) to disable KeepAlive. I setuped Apache mod\_proxy to disable KeepAlive by folowing settings. ``` ProxyPass /es http://localhost:9200 ProxyPassReverse /es http://localhost:9200 SetEnv proxy-nokeepalive 1 ``` And, I changed Kibana config.js to connect to ES via proxy. After that, the CPU usage dramatically came down. There may be some problem with HTTP KeepAlive at least on Solaris. Thanks, K.Cima ---
**Author:** ![jprante](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/jprante/32/44941_2.png) [@jprante](https://discuss.elastic.co/u/jprante)\ **Post date:** [February 19, 2014, 1:31pm UTC](https://discuss.elastic.co/t/very-high-sys-cpu-usage-with-http-keepalive/15895/2 "2014-02-19T13:31:46Z")
I think you do not need a proxy. Just disable TCP keep alive in HTTP. It is enabled by default. > **[Elasticsearch Platform — Find real-time answers at scale](https://www.elastic.co)** > > Power insights and outcomes with the Elasticsearch Platform and AI. See into your data and find answers that matter with enterprise solutions designed to help you build, observe, and protect. Try Elasticsearch free today. http.tcp\_keep\_alive: false Jörg On Wed, Feb 19, 2014 at 1:38 PM, K.Cima [shima@shakemid.com](mailto:shima@shakemid.com) wrote: > Hello, > > I am using Elasticsearch 1.0.0 with Kibana 3 on Solaris 11.1. My > environment > is as following. > > $ ./elasticsearch -v > Version: 1.0.0, Build: a46900e/2014-02-12T16:18:34Z, JVM: 1.7.0\_07 > > $ java -version > java version "1.7.0\_07" > Java(TM) SE Runtime Environment (build 1.7.0\_07-b10) > Java HotSpot(TM) Server VM (build 23.3-b01, mixed mode) > > $ cat /etc/release > Oracle Solaris 11.1 X86 > Copyright (c) 1983, 2012, Oracle and/or its affiliates. All rights > reserved. > Assembled 19 September 2012 > > I was suffering from very high CPU usage of Java process. > When I connected to ES via Kibana, the CPU usage became 90%-100%. > Strangely, > user cpu usage was not so high, but sys cpu usage was very high with vmstat > command. > > I tried to examine system calls by using truss command. I found there are > so > many ioctl calls while connecting. > > $ sudo truss -c -p 18496 > syscall seconds calls errors > read .000 13 4 > write .000 12 > ioctl 1.043 142731 > lwp\_cond\_wait .002 178 153 > lwp\_cond\_signal .000 25 > pollsys .000 54 > lwp\_mutex\_timedlock .000 1 > pread64 .016 1432 > -------- ------ ---- > sys totals: 1.062 144446 157 > usr time: .387 > elapsed: 3.620 > > $ sudo truss -p 18496 > ... > /47: ioctl(99, DP\_POLL, 0xF6CB5750) = 1 > /48: ioctl(102, DP\_POLL, 0xF6C746D0) = 1 > /49: ioctl(105, DP\_POLL, 0xF6C33650) = 1 > /48: ioctl(102, DP\_POLL, 0xF6C746D0) = 1 > /47: ioctl(99, DP\_POLL, 0xF6CB5750) = 1 > /48: ioctl(102, DP\_POLL, 0xF6C746D0) = 1 > /47: ioctl(99, DP\_POLL, 0xF6CB5750) = 1 > /49: ioctl(105, DP\_POLL, 0xF6C33650) = 1 > /47: ioctl(99, DP\_POLL, 0xF6CB5750) = 1 > /48: ioctl(102, DP\_POLL, 0xF6C746D0) = 1 > /49: ioctl(105, DP\_POLL, 0xF6C33650) = 1 > /47: ioctl(99, DP\_POLL, 0xF6CB5750) = 1 > /48: ioctl(102, DP\_POLL, 0xF6C746D0) = 1 > /49: ioctl(105, DP\_POLL, 0xF6C33650) = 1 > /47: ioctl(99, DP\_POLL, 0xF6CB5750) = 1 > /48: ioctl(102, DP\_POLL, 0xF6C746D0) = 1 > /47: ioctl(99, DP\_POLL, 0xF6CB5750) = 1 > ... > > I thought there are something wrong with HTTP KeepAlive. And, I considered > to put HTTP Proxy between ES and Kibana(client browser) to disable > KeepAlive. I setuped Apache mod\_proxy to disable KeepAlive by folowing > settings. > > ``` > ProxyPass /es http://localhost:9200 > ProxyPassReverse /es http://localhost:9200 > SetEnv proxy-nokeepalive 1 > > ``` > > And, I changed Kibana config.js to connect to ES via proxy. After that, the > CPU usage dramatically came down. There may be some problem with HTTP > KeepAlive at least on Solaris. > > Thanks, > K.Cima > > -- > View this message in context: > [http://elasticsearch-users.115913.n3.nabble.com/Very-high-sys-cpu-usage-with-HTTP-KeepAlive-tp4049998.html](http://elasticsearch-users.115913.n3.nabble.com/Very-high-sys-cpu-usage-with-HTTP-KeepAlive-tp4049998.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). > To view this discussion on the web visit > [https://groups.google.com/d/msgid/elasticsearch/1392813529284-4049998.post%40n3.nabble.com](https://groups.google.com/d/msgid/elasticsearch/1392813529284-4049998.post%40n3.nabble.com) > . > For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). -- 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). To view this discussion on the web visit [https://groups.google.com/d/msgid/elasticsearch/CAKdsXoHnjhFMbEsJxsT0HioQ1whvtXx77cfsr4U5Y%2BwzmyybXw%40mail.gmail.com](https://groups.google.com/d/msgid/elasticsearch/CAKdsXoHnjhFMbEsJxsT0HioQ1whvtXx77cfsr4U5Y%2BwzmyybXw%40mail.gmail.com). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![K\_Cima](https://avatars.discourse-cdn.com/v4/letter/k/58956e/32.png) [@K\_Cima](https://discuss.elastic.co/u/K_Cima)\ **Post date:** [February 19, 2014, 2:35pm UTC](https://discuss.elastic.co/t/very-high-sys-cpu-usage-with-http-keepalive/15895/3 "2014-02-19T14:35:10Z")
Thank you for the information. I tried both following settings in elasticsearch.yml right away. http.tcp\_keep\_alive: false network.tcp.keep\_alive: false But, unfortunately it seemed to have no effect. The cpu usage became high again. K.Cima ---
**Author:** ![Tony\_Su](https://avatars.discourse-cdn.com/v4/letter/t/8dc957/32.png) [@Tony\_Su](https://discuss.elastic.co/u/Tony_Su)\ **Post date:** [February 19, 2014, 4:05pm UTC](https://discuss.elastic.co/t/very-high-sys-cpu-usage-with-http-keepalive/15895/4 "2014-02-19T16:05:51Z")
Just a thought, What do you have pre-configured in Kibana? If you have some dashboards which are pulling large amounts of data at small time intervals, it might have a big effect. Also, I've found that the webserver you use might have a large effect. I haven't tested running Kibana as a plugin compared to running on Apache, but I know when I fire up any elasticsearch related web tool on Apache, there is a large effect on my systems(interestingly, if Apache is idle and no one is actively running a web tool there is no significant load). Tony On Wednesday, February 19, 2014 6:35:10 AM UTC-8, K.Cima wrote: > Thank you for the information. > I tried both following settings in elasticsearch.yml right away. > > http.tcp\_keep\_alive: false > network.tcp.keep\_alive: false > > But, unfortunately it seemed to have no effect. The cpu usage became high > again. > > K.Cima > > -- > View this message in context: > [http://elasticsearch-users.115913.n3.nabble.com/Very-high-sys-cpu-usage-with-HTTP-KeepAlive-tp4049998p4050010.html](http://elasticsearch-users.115913.n3.nabble.com/Very-high-sys-cpu-usage-with-HTTP-KeepAlive-tp4049998p4050010.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). To view this discussion on the web visit [https://groups.google.com/d/msgid/elasticsearch/197d6a4a-8caf-413e-834c-1fbde4901374%40googlegroups.com](https://groups.google.com/d/msgid/elasticsearch/197d6a4a-8caf-413e-834c-1fbde4901374%40googlegroups.com). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![K\_Cima](https://avatars.discourse-cdn.com/v4/letter/k/58956e/32.png) [@K\_Cima](https://discuss.elastic.co/u/K_Cima)\ **Post date:** [February 20, 2014, 10:29am UTC](https://discuss.elastic.co/t/very-high-sys-cpu-usage-with-http-keepalive/15895/5 "2014-02-20T10:29:10Z")
Thank you for your feedback. I am using Kibana with default config. And, I am testing ES with one index contains about only 1MB of documents. So, I don't think data size or queries from Kibana cause high cpu load in my case. K.Cima ---
**Author:** ![Tony\_Su](https://avatars.discourse-cdn.com/v4/letter/t/8dc957/32.png) [@Tony\_Su](https://discuss.elastic.co/u/Tony_Su)\ **Post date:** [February 20, 2014, 4:19pm UTC](https://discuss.elastic.co/t/very-high-sys-cpu-usage-with-http-keepalive/15895/6 "2014-02-20T16:19:08Z")
To clearly differentiate the Kibana from what it may be monitoring, Maybe deploy Kibana on a completely separate machine? That way, anything related to deploying Kibana can be isolated and studied. Local sensors (eg Marvel) and other processes not strictly part of Kibana but may be invoked by Kibana remain on the ES nodes. And, as I suggested, maybe launching on a different webserver engine might make a diff. Tony On Thursday, February 20, 2014 2:29:11 AM UTC-8, K.Cima wrote: > Thank you for your feedback. > I am using Kibana with default config. And, I am testing ES with one index > contains about only 1MB of documents. > So, I don't think data size or queries from Kibana cause high cpu load in > my > case. > > K.Cima > > -- > View this message in context: > [http://elasticsearch-users.115913.n3.nabble.com/Very-high-sys-cpu-usage-with-HTTP-KeepAlive-tp4049998p4050084.html](http://elasticsearch-users.115913.n3.nabble.com/Very-high-sys-cpu-usage-with-HTTP-KeepAlive-tp4049998p4050084.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). To view this discussion on the web visit [https://groups.google.com/d/msgid/elasticsearch/b630e573-77f3-4a7e-9765-82a2a77bb6d2%40googlegroups.com](https://groups.google.com/d/msgid/elasticsearch/b630e573-77f3-4a7e-9765-82a2a77bb6d2%40googlegroups.com). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![Binh\_Ly](https://avatars.discourse-cdn.com/v4/letter/b/ce7236/32.png) [@Binh\_Ly](https://discuss.elastic.co/u/Binh_Ly)\ **Post date:** [February 20, 2014, 4:24pm UTC](https://discuss.elastic.co/t/very-high-sys-cpu-usage-with-http-keepalive/15895/7 "2014-02-20T16:24:36Z")
Also, not sure exactly what caused your problem but these may help: 1. Tells you what threads are doing what in ES: curl "localhost:9200/\_nodes/hot\_threads" 1. Tells you how many http connections have been opened so far: curl "localhost:9200/\_nodes/stats/http?pretty" If that HTTP total count keeps on increasing very fast, that means HTTP keep-alive is not working somewhere in the chain between Kibana to ES. -- 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). To view this discussion on the web visit [https://groups.google.com/d/msgid/elasticsearch/75a94d2f-c957-4433-9aa5-a78ee227dd4e%40googlegroups.com](https://groups.google.com/d/msgid/elasticsearch/75a94d2f-c957-4433-9aa5-a78ee227dd4e%40googlegroups.com). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![K\_Cima](https://avatars.discourse-cdn.com/v4/letter/k/58956e/32.png) [@K\_Cima](https://discuss.elastic.co/u/K_Cima)\ **Post date:** [February 21, 2014, 12:42pm UTC](https://discuss.elastic.co/t/very-high-sys-cpu-usage-with-http-keepalive/15895/8 "2014-02-21T12:42:24Z")
Thank you for your reply. I installed Kibana on separated host to eliminate effect of webserver. But, the CPU usage of ES host became high just like before. The hot thread seems to be http\_server\_worker. And, total\_opened connections did not increase while repeatedly connecting from Kibana. So, I think KeepAlive does work, but there seems to be something wrong with KeepAlive. There may be some problems between HTTP KeepAlive implementation and Solaris /dev/poll. [http://server:9200/\_nodes/hot\_threads](http://server:9200/_nodes/hot_threads) ::: [Tom Corsi][OOF\_Kl3qS82C0GuJX-nzdA][server][inet[/192.168.0.3:9300]] 100.7% (503.6ms out of 500ms) cpu usage by thread 'elasticsearch[Tom Corsi][http\_server\_worker][T#8]{New I/O worker #26}' 10/10 snapshots sharing following 15 elements sun.nio.ch.DevPollArrayWrapper.poll0(Native Method) sun.nio.ch.DevPollArrayWrapper.poll(DevPollArrayWrapper.java:220) sun.nio.ch.DevPollSelectorImpl.doSelect(DevPollSelectorImpl.java:84) sun.nio.ch.SelectorImpl.lockAndDoSelect(SelectorImpl.java:87) sun.nio.ch.SelectorImpl.select(SelectorImpl.java:98) org.elasticsearch.common.netty.channel.socket.nio.SelectorUtil.select(SelectorUtil.java:68) org.elasticsearch.common.netty.channel.socket.nio.AbstractNioSelector.select(AbstractNioSelector.java:415) org.elasticsearch.common.netty.channel.socket.nio.AbstractNioSelector.run(AbstractNioSelector.java:212) org.elasticsearch.common.netty.channel.socket.nio.AbstractNioWorker.run(AbstractNioWorker.java:89) org.elasticsearch.common.netty.channel.socket.nio.NioWorker.run(NioWorker.java:178) org.elasticsearch.common.netty.util.ThreadRenamingRunnable.run(ThreadRenamingRunnable.java:108) org.elasticsearch.common.netty.util.internal.DeadLockProofWorker$1.run(DeadLockProofWorker.java:42) java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1110) java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:603) java.lang.Thread.run(Thread.java:722) 100.7% (503.3ms out of 500ms) cpu usage by thread 'elasticsearch[Tom Corsi][http\_server\_worker][T#3]{New I/O worker #21}' 10/10 snapshots sharing following 15 elements sun.nio.ch.DevPollArrayWrapper.poll0(Native Method) sun.nio.ch.DevPollArrayWrapper.poll(DevPollArrayWrapper.java:220) sun.nio.ch.DevPollSelectorImpl.doSelect(DevPollSelectorImpl.java:84) sun.nio.ch.SelectorImpl.lockAndDoSelect(SelectorImpl.java:87) sun.nio.ch.SelectorImpl.select(SelectorImpl.java:98) org.elasticsearch.common.netty.channel.socket.nio.SelectorUtil.select(SelectorUtil.java:68) org.elasticsearch.common.netty.channel.socket.nio.AbstractNioSelector.select(AbstractNioSelector.java:415) org.elasticsearch.common.netty.channel.socket.nio.AbstractNioSelector.run(AbstractNioSelector.java:212) org.elasticsearch.common.netty.channel.socket.nio.AbstractNioWorker.run(AbstractNioWorker.java:89) org.elasticsearch.common.netty.channel.socket.nio.NioWorker.run(NioWorker.java:178) org.elasticsearch.common.netty.util.ThreadRenamingRunnable.run(ThreadRenamingRunnable.java:108) org.elasticsearch.common.netty.util.internal.DeadLockProofWorker$1.run(DeadLockProofWorker.java:42) java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1110) java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:603) java.lang.Thread.run(Thread.java:722) 100.7% (503.3ms out of 500ms) cpu usage by thread 'elasticsearch[Tom Corsi][http\_server\_worker][T#2]{New I/O worker #20}' 10/10 snapshots sharing following 15 elements sun.nio.ch.DevPollArrayWrapper.poll0(Native Method) sun.nio.ch.DevPollArrayWrapper.poll(DevPollArrayWrapper.java:220) sun.nio.ch.DevPollSelectorImpl.doSelect(DevPollSelectorImpl.java:84) sun.nio.ch.SelectorImpl.lockAndDoSelect(SelectorImpl.java:87) sun.nio.ch.SelectorImpl.select(SelectorImpl.java:98) org.elasticsearch.common.netty.channel.socket.nio.SelectorUtil.select(SelectorUtil.java:68) org.elasticsearch.common.netty.channel.socket.nio.AbstractNioSelector.select(AbstractNioSelector.java:415) org.elasticsearch.common.netty.channel.socket.nio.AbstractNioSelector.run(AbstractNioSelector.java:212) org.elasticsearch.common.netty.channel.socket.nio.AbstractNioWorker.run(AbstractNioWorker.java:89) org.elasticsearch.common.netty.channel.socket.nio.NioWorker.run(NioWorker.java:178) org.elasticsearch.common.netty.util.ThreadRenamingRunnable.run(ThreadRenamingRunnable.java:108) org.elasticsearch.common.netty.util.internal.DeadLockProofWorker$1.run(DeadLockProofWorker.java:42) java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1110) java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:603) java.lang.Thread.run(Thread.java:722) [http://server:9200/\_nodes/stats/http?pretty](http://server:9200/_nodes/stats/http?pretty) { "cluster\_name" : "elasticsearch", "nodes" : { "cfkkA--DTCa1oOW2LDHomg" : { "timestamp" : 1392985424001, "name" : "Volpan", "transport\_address" : "inet[/192.168.0.3:9300]", "host" : "server", "ip" : ["inet[/192.168.0.3:9300]", "NONE" ], "http" : { "current\_open" : 4, "total\_opened" : 5 } } } } ---
**Author:** ![K\_Cima](https://avatars.discourse-cdn.com/v4/letter/k/58956e/32.png) [@K\_Cima](https://discuss.elastic.co/u/K_Cima)\ **Post date:** [February 21, 2014, 2:26pm UTC](https://discuss.elastic.co/t/very-high-sys-cpu-usage-with-http-keepalive/15895/9 "2014-02-21T14:26:36Z")
> I found similar issue about Netty. I updated Java version 1.7.0\_07 to 1.7.0\_51. After that, the heavy CPU load has gone. It seemed to be bug of Java. ---
**Author:** ![jprante](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/jprante/32/44941_2.png) [@jprante](https://discuss.elastic.co/u/jprante)\ **Post date:** [February 23, 2014, 12:48pm UTC](https://discuss.elastic.co/t/very-high-sys-cpu-usage-with-http-keepalive/15895/10 "2014-02-23T12:48:37Z")
Yes, it is a known Java "epoll" bug that is fixed in Java \>= 7u11 Jörg -- 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). To view this discussion on the web visit [https://groups.google.com/d/msgid/elasticsearch/CAKdsXoHfW4HKLnAFhopX66dcMz%3DbiewbkEPPwF0NrGdSTHoLXg%40mail.gmail.com](https://groups.google.com/d/msgid/elasticsearch/CAKdsXoHfW4HKLnAFhopX66dcMz%3DbiewbkEPPwF0NrGdSTHoLXg%40mail.gmail.com). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![K\_Cima](https://avatars.discourse-cdn.com/v4/letter/k/58956e/32.png) [@K\_Cima](https://discuss.elastic.co/u/K_Cima)\ **Post date:** [February 24, 2014, 4:27am UTC](https://discuss.elastic.co/t/very-high-sys-cpu-usage-with-http-keepalive/15895/11 "2014-02-24T04:27:35Z")
All right. Thank you! ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [July 6, 2017, 1:48am UTC](https://discuss.elastic.co/t/very-high-sys-cpu-usage-with-http-keepalive/15895/12 "2017-07-06T01:48:00Z")