# HIgh CPU usage without processing any data
**URL:** https://discuss.elastic.co/t/high-cpu-usage-without-processing-any-data/199147
**Category:** Logstash
**Created:** [September 11, 2019, 9:06pm UTC](https://discuss.elastic.co/t/high-cpu-usage-without-processing-any-data/199147 "2019-09-11T21:06:37Z")
**Posts on this page:** 4
**Page:** 1
### Author:  [@stefws](https://discuss.elastic.co/u/stefws)
#### Post date: [September 11, 2019, 9:06pm UTC](https://discuss.elastic.co/t/high-cpu-usage-without-processing-any-data/199147/1 "2019-09-11T21:06:37Z")
Trying to config a new logstash instance to handle syslog input, so prepared input, output and a dummy working filter as a starting template from another running instance, but not sending any data to input yet. Even so logstash is burning CPU in one thread #15 running bootstrap environment, wonder why and how to stop this useless CO2 generation 🙂
> $ curl -XGET '[http://localhost:9600/\_node/hot\_threads?pretty](http://localhost:9600/_node/hot_threads?pretty)'
> {
> "host" : "",
> "version" : "6.8.3",
> "http\_address" : "127.0.0.1:9600",
> "id" : "a6cc457a-1610-4e8d-9309-77ce10c865ab",
> "name" : "",
> "hot\_threads" : {
> "time" : "2019-09-11T22:59:14+02:00",
> "busiest\_threads" : 10,
> "threads" : [ {
> "name" : "Ruby-0-Thread-1",
> "thread\_id" : 15,
> "percent\_of\_cpu\_time" : 96.31,
> "state" : "runnable",
> "path" : "/usr/share/logstash/lib/bootstrap/environment.rb:6",
> "traces" : ["java.lang.System.identityHashCode(Native Method)", "java.util.IdentityHashMap.hash(Unknown Source)", "java.util.IdentityHashMap.containsKey(Unknown Source)"]
> }, {
> "name" : "LogStash::Runner",
> "thread\_id" : 1,
> "percent\_of\_cpu\_time" : 1.2,
> "state" : "timed\_waiting",
> "traces" : ["java.lang.Object.wait(Native Method)", "java.lang.Thread.join(Unknown Source)", "org.jruby.internal.runtime.NativeThread.join(NativeThread.java:75)"]
> }, {
> "name" : "[main]\>worker7",
> "thread\_id" : 36,
> "percent\_of\_cpu\_time" : 0.25,
> "state" : "timed\_waiting",
> "traces" : ["sun.misc.Unsafe.park(Native Method)", "java.util.concurrent.locks.LockSupport.parkNanos(Unknown Source)", "java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject.awaitNanos(Unknown Source)"]
> }, {
> ...
Nothing shows in the log than the normal start up:
> [2019-09-11T22:39:38,662][INFO][logstash.pipeline] Pipeline started successfully {:pipeline\_id=\>"main", :thread=\>"#\"}
> [2019-09-11T22:39:38,709][INFO][logstash.agent] Pipelines running {:count=\>1, :running\_pipelines=\>[:main], :non\_running\_pipelines=\>}
> [2019-09-11T22:39:39,119][INFO][logstash.agent] Successfully started Logstash API endpoint {:port=\>9600}
---
### Author:  [@stefws](https://discuss.elastic.co/u/stefws)
#### Post date: [September 13, 2019, 8:51am UTC](https://discuss.elastic.co/t/high-cpu-usage-without-processing-any-data/199147/2 "2019-09-13T08:51:50Z")
Any one got an idea how to attack this...
---
### Author:  [@stefws](https://discuss.elastic.co/u/stefws)
#### Post date: [September 13, 2019, 8:56pm UTC](https://discuss.elastic.co/t/high-cpu-usage-without-processing-any-data/199147/3 "2019-09-13T20:56:30Z")
Hm it this just the norm, got other working logstash instances, they also seem to burn one cpu core near 100% in a Ruby-0-Thread... seems like some kind of busy-waiting thread
Top -H from a non-event-processing instance:
Tasks: 425 total, 2 running, 324 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.1%us, 1.0%sy, 6.4%ni, 92.4%id, 0.0%wa, 0.0%hi, 0.0%si, 0.1%st
Mem: 16424228k total, 4932720k used, 11491508k free, 18400k buffers
Swap: 3321852k total, 0k used, 3321852k free, 265600k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
3358 logstash 39 19 9.9g 1.7g 19m R 98.3 10.6 1:04.03 LogStash::Runne
3370 logstash 39 19 9.9g 1.7g 19m S 1.3 10.6 0:00.53 [main]\>worker0
3372 logstash 39 19 9.9g 1.7g 19m S 1.3 10.6 0:00.48 [main]\>worker2
3376 logstash 39 19 9.9g 1.7g 19m S 1.3 10.6 0:00.54 [main]\>worker6
3336 logstash 39 19 9.9g 1.7g 19m S 1.0 10.6 0:01.47 java
3337 logstash 39 19 9.9g 1.7g 19m S 1.0 10.6 0:00.64 java
3371 logstash 39 19 9.9g 1.7g 19m S 1.0 10.6 0:00.50 [main]\>worker1
3373 logstash 39 19 9.9g 1.7g 19m S 1.0 10.6 0:00.52 [main]\>worker3
3375 logstash 39 19 9.9g 1.7g 19m S 1.0 10.6 0:00.54 [main]\>worker5
3318 logstash 39 19 9.9g 1.7g 19m S 0.7 10.6 0:00.63 java
3322 logstash 39 19 9.9g 1.7g 19m S 0.7 10.6 0:00.61 java
something most be going on in Userland as strace only shows this:
[root@es-i1 ~]# ps -fulogstash
UID PID PPID C STIME TTY TIME CMD
logstash 3303 1 99 Sep13 ? 2-16:33:40 /usr/bin/java -Xms4g -Xmx4g
[root@es-i1 ~]# strace -p 3303
Process 3303 attached
futex(0x7ff4c6add9d0, FUTEX\_WAIT, 3317, NULL^CProcess 3303 detached
\
---
### Author:  [@system](https://discuss.elastic.co/u/system)
#### Post date: [October 11, 2019, 8:56pm UTC](https://discuss.elastic.co/t/high-cpu-usage-without-processing-any-data/199147/4 "2019-10-11T20:56:32Z")
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.