# Help with GC configuration **URL:** \ **Category:** Elasticsearch\ **Created:** [August 27, 2013, 1:45am UTC](https://discuss.elastic.co/t/help-with-gc-configuration/13339 "2013-08-27T01:45:29Z")\ **Posts on this page:** 6\ **Page:** 1
**Author:** ![alucero](https://avatars.discourse-cdn.com/v4/letter/a/839c29/32.png) [@alucero](https://discuss.elastic.co/u/alucero)\ **Post date:** [August 27, 2013, 1:45am UTC](https://discuss.elastic.co/t/help-with-gc-configuration/13339/1 "2013-08-27T01:45:29Z")
Hello, we are experimenting some problems with GC configuration of ElasticSearch, we have elasticsearch 0.20.2 running in java 6.0.32 with 9 node with 32 gb and 16 gb assigned to the elastic, in this cluster, we have one index with 20 shard and 2 replicas, we have around 15k of request per minute. After a few hours, at least one node is expelled of the cluster and the logs that we see are: [2013-08-26 20:33:36,030][INFO][monitor.jvm] [King Bedlam] [gc][ConcurrentMarkSweep][16432][221] duration [18.9s], collections [2]/[2.2m], total [18.9s]/[1.2m], memory [6.8gb]-\>[6.6gb]/[6.9gb], all\_pools {[Code Cache] [5.9mb]-\>[5.9mb]/[48mb]}{[Par Eden Space] [91.7mb]-\>[611.4kb]/[133.1mb]}{[Par Survivor Space] [16.6mb]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.7gb]-\>[6.6gb]/[6.8gb]}{[CMS Perm Gen] [42.6mb]-\>[42.2mb]/[82mb]} [2013-08-26 20:36:11,033][WARN][monitor.jvm] [King Bedlam] [gc][ConcurrentMarkSweep][16439][223] duration [31.3s], collections [2]/[2.4m], total [31.3s]/[1.7m], memory [6.8gb]-\>[6.7gb]/[6.9gb], all\_pools {[Code Cache] [5.9mb]-\>[5.9mb]/[48mb]}{[Par Eden Space] [68.8mb]-\>[32mb]/[133.1mb]}{[Par Survivor Space] [16.6mb]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.7gb]-\>[6.7gb]/[6.8gb]}{[CMS Perm Gen] [42.2mb]-\>[42.2mb]/[82mb]} [2013-08-26 20:38:54,467][WARN][monitor.jvm] [King Bedlam] [gc][ConcurrentMarkSweep][16441][225] duration [37.8s], collections [2]/[2.7m], total [37.8s]/[2.4m], memory [6.8gb]-\>[6.8gb]/[6.9gb], all\_pools {[Code Cache] [6mb]-\>[6mb]/[48mb]}{[Par Eden Space] [83.5mb]-\>[1mb]/[133.1mb]}{[Par Survivor Space] [0b]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.7gb]-\>[6.8gb]/[6.8gb]}{[CMS Perm Gen] [42.2mb]-\>[42.2mb]/[82mb]} [2013-08-26 20:41:36,086][WARN][monitor.jvm] [King Bedlam] [gc][ConcurrentMarkSweep][16442][227] duration [42.1s], collections [2]/[2.6m], total [42.1s]/[3.1m], memory [6.8gb]-\>[6.8gb]/[6.9gb], all\_pools {[Code Cache] [6mb]-\>[6mb]/[48mb]}{[Par Eden Space] [1mb]-\>[9.8mb]/[133.1mb]}{[Par Survivor Space] [0b]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.8gb]-\>[6.8gb]/[6.8gb]}{[CMS Perm Gen] [42.2mb]-\>[42.2mb]/[82mb]} [2013-08-26 20:44:23,710][WARN][monitor.jvm] [King Bedlam] [gc][ConcurrentMarkSweep][16445][229] duration [40.1s], collections [2]/[2.7m], total [40.1s]/[3.7m], memory [6.9gb]-\>[6.8gb]/[6.9gb], all\_pools {[Code Cache] [6mb]-\>[6mb]/[48mb]}{[Par Eden Space] [115.4mb]-\>[56.4mb]/[133.1mb]}{[Par Survivor Space] [0b]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.8gb]-\>[6.8gb]/[6.8gb]}{[CMS Perm Gen] [42.2mb]-\>[42.2mb]/[82mb]} I don't know how we can resolve this problem, we try diferents configurations in the in the elasticsearch.in.sh but the problem persist. would be great if someone could help with this. regards -- 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). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![Mohit\_Anchlia](https://avatars.discourse-cdn.com/v4/letter/m/848f3c/32.png) [@Mohit\_Anchlia](https://discuss.elastic.co/u/Mohit_Anchlia)\ **Post date:** [August 27, 2013, 2:12am UTC](https://discuss.elastic.co/t/help-with-gc-configuration/13339/2 "2013-08-27T02:12:06Z")
How many cores do you have? Do you see a CPU spike during gc? Can you first test with 12 gb heap? Also, I suggest increasing your Eden space to 2 gb After that enable gc logging and upload new information. Sent from my iPhone On Aug 26, 2013, at 6:45 PM, Ariel L [ariel2129@gmail.com](mailto:ariel2129@gmail.com) wrote: > Hello, we are experimenting some problems with GC configuration of Elasticsearch, > > we have elasticsearch 0.20.2 running in java 6.0.32 with 9 node with 32 gb and 16 gb assigned to the elastic, in this cluster, we have one index with 20 shard and 2 replicas, we have around 15k of request per minute. > > After a few hours, at least one node is expelled of the cluster and the logs that we see are: > > [2013-08-26 20:33:36,030][INFO][monitor.jvm] [King Bedlam] [gc][ConcurrentMarkSweep][16432][221] duration [18.9s], collections [2]/[2.2m], total [18.9s]/[1.2m], memory [6.8gb]-\>[6.6gb]/[6.9gb], all\_pools {[Code Cache] [5.9mb]-\>[5.9mb]/[48mb]}{[Par Eden Space] [91.7mb]-\>[611.4kb]/[133.1mb]}{[Par Survivor Space] [16.6mb]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.7gb]-\>[6.6gb]/[6.8gb]}{[CMS Perm Gen] [42.6mb]-\>[42.2mb]/[82mb]} > > [2013-08-26 20:36:11,033][WARN][monitor.jvm] [King Bedlam] [gc][ConcurrentMarkSweep][16439][223] duration [31.3s], collections [2]/[2.4m], total [31.3s]/[1.7m], memory [6.8gb]-\>[6.7gb]/[6.9gb], all\_pools {[Code Cache] [5.9mb]-\>[5.9mb]/[48mb]}{[Par Eden Space] [68.8mb]-\>[32mb]/[133.1mb]}{[Par Survivor Space] [16.6mb]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.7gb]-\>[6.7gb]/[6.8gb]}{[CMS Perm Gen] [42.2mb]-\>[42.2mb]/[82mb]} > > [2013-08-26 20:38:54,467][WARN][monitor.jvm] [King Bedlam] [gc][ConcurrentMarkSweep][16441][225] duration [37.8s], collections [2]/[2.7m], total [37.8s]/[2.4m], memory [6.8gb]-\>[6.8gb]/[6.9gb], all\_pools {[Code Cache] [6mb]-\>[6mb]/[48mb]}{[Par Eden Space] [83.5mb]-\>[1mb]/[133.1mb]}{[Par Survivor Space] [0b]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.7gb]-\>[6.8gb]/[6.8gb]}{[CMS Perm Gen] [42.2mb]-\>[42.2mb]/[82mb]} > > [2013-08-26 20:41:36,086][WARN][monitor.jvm] [King Bedlam] [gc][ConcurrentMarkSweep][16442][227] duration [42.1s], collections [2]/[2.6m], total [42.1s]/[3.1m], memory [6.8gb]-\>[6.8gb]/[6.9gb], all\_pools {[Code Cache] [6mb]-\>[6mb]/[48mb]}{[Par Eden Space] [1mb]-\>[9.8mb]/[133.1mb]}{[Par Survivor Space] [0b]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.8gb]-\>[6.8gb]/[6.8gb]}{[CMS Perm Gen] [42.2mb]-\>[42.2mb]/[82mb]} > > [2013-08-26 20:44:23,710][WARN][monitor.jvm] [King Bedlam] [gc][ConcurrentMarkSweep][16445][229] duration [40.1s], collections [2]/[2.7m], total [40.1s]/[3.7m], memory [6.9gb]-\>[6.8gb]/[6.9gb], all\_pools {[Code Cache] [6mb]-\>[6mb]/[48mb]}{[Par Eden Space] [115.4mb]-\>[56.4mb]/[133.1mb]}{[Par Survivor Space] [0b]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.8gb]-\>[6.8gb]/[6.8gb]}{[CMS Perm Gen] [42.2mb]-\>[42.2mb]/[82mb]} > > I don't know how we can resolve this problem, we try diferents configurations in the in the elasticsearch.in.sh but the problem persist. > > would be great if someone could help with this. > regards > > -- > 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). > 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). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![Runar\_Myklebust](https://avatars.discourse-cdn.com/v4/letter/r/a6a055/32.png) [@Runar\_Myklebust](https://discuss.elastic.co/u/Runar_Myklebust)\ **Post date:** [August 27, 2013, 10:18am UTC](https://discuss.elastic.co/t/help-with-gc-configuration/13339/3 "2013-08-27T10:18:41Z")
In addition to the ConcurrentMarkSweep, consider tuning these options also: -XX:CMSInitiatingOccupancyFraction=30 -XX:+UseCMSInitiatingOccupancyOnly -XX:NewRatio=4 The actual values here may vary, but this is a configuration working well at some of out customers. Also, I dont know what difference there are between v.0.19.10 and 0.20.2 vs 0.20.6, but we had issues in 0.19.10 with long GC that worked far better in 0.20.6. On Tue, Aug 27, 2013 at 4:12 AM, Mohit Anchlia [mohitanchlia@gmail.com](mailto:mohitanchlia@gmail.com)wrote: > How many cores do you have? Do you see a CPU spike during gc? > > Can you first test with 12 gb heap? > > Also, I suggest increasing your Eden space to 2 gb > > After that enable gc logging and upload new information. > > Sent from my iPhone > > On Aug 26, 2013, at 6:45 PM, Ariel L [ariel2129@gmail.com](mailto:ariel2129@gmail.com) wrote: > > Hello, we are experimenting some problems with GC configuration of > Elasticsearch, > > we have elasticsearch 0.20.2 running in java 6.0.32 with 9 node with 32 gb > and 16 gb assigned to the elastic, in this cluster, we have one index with > 20 shard and 2 replicas, we have around 15k of request per minute. > > After a few hours, at least one node is expelled of the cluster and the > logs that we see are: > > [2013-08-26 20:33:36,030][INFO][monitor.jvm] [King Bedlam] > [gc][ConcurrentMarkSweep][16432][221] duration [18.9s], collections > [2]/[2.2m], total [18.9s]/[1.2m], memory [6.8gb]-\>[6.6gb]/[6.9gb], > all\_pools {[Code Cache] [5.9mb]-\>[5.9mb]/[48mb]}{[Par Eden Space] > [91.7mb]-\>[611.4kb]/[133.1mb]}{[Par Survivor Space] > [16.6mb]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.7gb]-\>[6.6gb]/[6.8gb]}{[CMS Perm > Gen] [42.6mb]-\>[42.2mb]/[82mb]} > > [2013-08-26 20:36:11,033][WARN][monitor.jvm] [King Bedlam] > [gc][ConcurrentMarkSweep][16439][223] duration [31.3s], collections > [2]/[2.4m], total [31.3s]/[1.7m], memory [6.8gb]-\>[6.7gb]/[6.9gb], > all\_pools {[Code Cache] [5.9mb]-\>[5.9mb]/[48mb]}{[Par Eden Space] > [68.8mb]-\>[32mb]/[133.1mb]}{[Par Survivor Space] > [16.6mb]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.7gb]-\>[6.7gb]/[6.8gb]}{[CMS Perm > Gen] [42.2mb]-\>[42.2mb]/[82mb]} > > [2013-08-26 20:38:54,467][WARN][monitor.jvm] [King Bedlam] > [gc][ConcurrentMarkSweep][16441][225] duration [37.8s], collections > [2]/[2.7m], total [37.8s]/[2.4m], memory [6.8gb]-\>[6.8gb]/[6.9gb], > all\_pools {[Code Cache] [6mb]-\>[6mb]/[48mb]}{[Par Eden Space] > [83.5mb]-\>[1mb]/[133.1mb]}{[Par Survivor Space] [0b]-\>[0b]/[16.6mb]}{[CMS > Old Gen] [6.7gb]-\>[6.8gb]/[6.8gb]}{[CMS Perm Gen] [42.2mb]-\>[42.2mb]/[82mb]} > > [2013-08-26 20:41:36,086][WARN][monitor.jvm] [King Bedlam] > [gc][ConcurrentMarkSweep][16442][227] duration [42.1s], collections > [2]/[2.6m], total [42.1s]/[3.1m], memory [6.8gb]-\>[6.8gb]/[6.9gb], > all\_pools {[Code Cache] [6mb]-\>[6mb]/[48mb]}{[Par Eden Space] > [1mb]-\>[9.8mb]/[133.1mb]}{[Par Survivor Space] [0b]-\>[0b]/[16.6mb]}{[CMS > Old Gen] [6.8gb]-\>[6.8gb]/[6.8gb]}{[CMS Perm Gen] [42.2mb]-\>[42.2mb]/[82mb]} > > [2013-08-26 20:44:23,710][WARN][monitor.jvm] [King Bedlam] > [gc][ConcurrentMarkSweep][16445][229] duration [40.1s], collections > [2]/[2.7m], total [40.1s]/[3.7m], memory [6.9gb]-\>[6.8gb]/[6.9gb], > all\_pools {[Code Cache] [6mb]-\>[6mb]/[48mb]}{[Par Eden Space] > [115.4mb]-\>[56.4mb]/[133.1mb]}{[Par Survivor Space] > [0b]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.8gb]-\>[6.8gb]/[6.8gb]}{[CMS Perm Gen] > [42.2mb]-\>[42.2mb]/[82mb]} > > I don't know how we can resolve this problem, we try diferents > configurations in the in the elasticsearch.in.sh but the problem persist. > > would be great if someone could help with this. > > regards > > -- > 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). > 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). > For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ## -- mvh Runar Myklebust -- 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). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![Ivan](https://avatars.discourse-cdn.com/v4/letter/i/df788c/32.png) [@Ivan](https://discuss.elastic.co/u/Ivan)\ **Post date:** [August 27, 2013, 6:11pm UTC](https://discuss.elastic.co/t/help-with-gc-configuration/13339/4 "2013-08-27T18:11:37Z")
I am not want to upgrade my software stack regularly, but the memory improvements in elasticsearch 0.90 are excellent. I would spend time trying to upgrade elasticsearch instead of fine-tuning the JVM or upgrading to Java 7. Yes, they should still be tuned, but the benefits are not as great as upgrading elasticsearch. Cheers, Ivan On Tue, Aug 27, 2013 at 3:18 AM, Runar Myklebust [runar.a.m@gmail.com](mailto:runar.a.m@gmail.com)wrote: > In addition to the ConcurrentMarkSweep, consider tuning these options > also: > -XX:CMSInitiatingOccupancyFraction=30 > -XX:+UseCMSInitiatingOccupancyOnly > -XX:NewRatio=4 > > The actual values here may vary, but this is a configuration working well > at some of out customers. > Also, I dont know what difference there are between v.0.19.10 and 0.20.2 > vs 0.20.6, but we had issues in 0.19.10 with long GC that worked far better > in 0.20.6. > > On Tue, Aug 27, 2013 at 4:12 AM, Mohit Anchlia [mohitanchlia@gmail.com](mailto:mohitanchlia@gmail.com)wrote: > > > How many cores do you have? Do you see a CPU spike during gc? > > > > Can you first test with 12 gb heap? > > > > Also, I suggest increasing your Eden space to 2 gb > > > > After that enable gc logging and upload new information. > > > > Sent from my iPhone > > > > On Aug 26, 2013, at 6:45 PM, Ariel L [ariel2129@gmail.com](mailto:ariel2129@gmail.com) wrote: > > > > Hello, we are experimenting some problems with GC configuration of > > Elasticsearch, > > > > we have elasticsearch 0.20.2 running in java 6.0.32 with 9 node with 32 > > gb and 16 gb assigned to the elastic, in this cluster, we have one index > > with 20 shard and 2 replicas, we have around 15k of request per minute. > > > > After a few hours, at least one node is expelled of the cluster and the > > logs that we see are: > > > > [2013-08-26 20:33:36,030][INFO][monitor.jvm] [King Bedlam] > > [gc][ConcurrentMarkSweep][16432][221] duration [18.9s], collections > > [2]/[2.2m], total [18.9s]/[1.2m], memory [6.8gb]-\>[6.6gb]/[6.9gb], > > all\_pools {[Code Cache] [5.9mb]-\>[5.9mb]/[48mb]}{[Par Eden Space] > > [91.7mb]-\>[611.4kb]/[133.1mb]}{[Par Survivor Space] > > [16.6mb]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.7gb]-\>[6.6gb]/[6.8gb]}{[CMS Perm > > Gen] [42.6mb]-\>[42.2mb]/[82mb]} > > > > [2013-08-26 20:36:11,033][WARN][monitor.jvm] [King Bedlam] > > [gc][ConcurrentMarkSweep][16439][223] duration [31.3s], collections > > [2]/[2.4m], total [31.3s]/[1.7m], memory [6.8gb]-\>[6.7gb]/[6.9gb], > > all\_pools {[Code Cache] [5.9mb]-\>[5.9mb]/[48mb]}{[Par Eden Space] > > [68.8mb]-\>[32mb]/[133.1mb]}{[Par Survivor Space] > > [16.6mb]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.7gb]-\>[6.7gb]/[6.8gb]}{[CMS Perm > > Gen] [42.2mb]-\>[42.2mb]/[82mb]} > > > > [2013-08-26 20:38:54,467][WARN][monitor.jvm] [King Bedlam] > > [gc][ConcurrentMarkSweep][16441][225] duration [37.8s], collections > > [2]/[2.7m], total [37.8s]/[2.4m], memory [6.8gb]-\>[6.8gb]/[6.9gb], > > all\_pools {[Code Cache] [6mb]-\>[6mb]/[48mb]}{[Par Eden Space] > > [83.5mb]-\>[1mb]/[133.1mb]}{[Par Survivor Space] [0b]-\>[0b]/[16.6mb]}{[CMS > > Old Gen] [6.7gb]-\>[6.8gb]/[6.8gb]}{[CMS Perm Gen] [42.2mb]-\>[42.2mb]/[82mb]} > > > > [2013-08-26 20:41:36,086][WARN][monitor.jvm] [King Bedlam] > > [gc][ConcurrentMarkSweep][16442][227] duration [42.1s], collections > > [2]/[2.6m], total [42.1s]/[3.1m], memory [6.8gb]-\>[6.8gb]/[6.9gb], > > all\_pools {[Code Cache] [6mb]-\>[6mb]/[48mb]}{[Par Eden Space] > > [1mb]-\>[9.8mb]/[133.1mb]}{[Par Survivor Space] [0b]-\>[0b]/[16.6mb]}{[CMS > > Old Gen] [6.8gb]-\>[6.8gb]/[6.8gb]}{[CMS Perm Gen] [42.2mb]-\>[42.2mb]/[82mb]} > > > > [2013-08-26 20:44:23,710][WARN][monitor.jvm] [King Bedlam] > > [gc][ConcurrentMarkSweep][16445][229] duration [40.1s], collections > > [2]/[2.7m], total [40.1s]/[3.7m], memory [6.9gb]-\>[6.8gb]/[6.9gb], > > all\_pools {[Code Cache] [6mb]-\>[6mb]/[48mb]}{[Par Eden Space] > > [115.4mb]-\>[56.4mb]/[133.1mb]}{[Par Survivor Space] > > [0b]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.8gb]-\>[6.8gb]/[6.8gb]}{[CMS Perm Gen] > > [42.2mb]-\>[42.2mb]/[82mb]} > > > > I don't know how we can resolve this problem, we try diferents > > configurations in the in the elasticsearch.in.sh but the problem > > persist. > > > > would be great if someone could help with this. > > > > regards > > > > -- > > 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). > > 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). > > For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). > > ## -- mvh > > Runar Myklebust > > -- > 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). > 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). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**Author:** ![Israel\_Ekpo](https://avatars.discourse-cdn.com/v4/letter/i/49beb7/32.png) [@Israel\_Ekpo](https://discuss.elastic.co/u/Israel_Ekpo)\ **Post date:** [August 27, 2013, 7:09pm UTC](https://discuss.elastic.co/t/help-with-gc-configuration/13339/5 "2013-08-27T19:09:10Z")
Take a look at this Webinar[http://www.elasticsearch.org/webinars/whats-new-with-elasticsearch-0-90/](http://www.elasticsearch.org/webinars/whats-new-with-elasticsearch-0-90/)in its entirety: > **[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. I don't think you will need much convincing after you thoroughly understand all the changes that have been made in Elasticsearch to leverage the mind-blowing performance improvements that comes with the new version of Lucene (the underlying Information Retrieval library that powers Elasticsearch). _Author and Instructor for the Upcoming Book and Lecture Series_ _Massive Log Data Aggregation, Processing, Searching and Visualization with Open Source Software_ _[http://massivelogdata.com](http://massivelogdata.com)_ On 27 August 2013 14:11, Ivan Brusic [ivan@brusic.com](mailto:ivan@brusic.com) wrote: > I am not want to upgrade my software stack regularly, but the memory > improvements in elasticsearch 0.90 are excellent. I would spend time trying > to upgrade elasticsearch instead of fine-tuning the JVM or upgrading to > Java 7. Yes, they should still be tuned, but the benefits are not as great > as upgrading elasticsearch. > > Cheers, > > Ivan > > On Tue, Aug 27, 2013 at 3:18 AM, Runar Myklebust [runar.a.m@gmail.com](mailto:runar.a.m@gmail.com)wrote: > > > In addition to the ConcurrentMarkSweep, consider tuning these options > > also: > > -XX:CMSInitiatingOccupancyFraction=30 > > -XX:+UseCMSInitiatingOccupancyOnly > > -XX:NewRatio=4 > > > > The actual values here may vary, but this is a configuration working well > > at some of out customers. > > Also, I dont know what difference there are between v.0.19.10 and 0.20.2 > > vs 0.20.6, but we had issues in 0.19.10 with long GC that worked far better > > in 0.20.6. > > > > On Tue, Aug 27, 2013 at 4:12 AM, Mohit Anchlia [mohitanchlia@gmail.com](mailto:mohitanchlia@gmail.com)wrote: > > > > > How many cores do you have? Do you see a CPU spike during gc? > > > > > > Can you first test with 12 gb heap? > > > > > > Also, I suggest increasing your Eden space to 2 gb > > > > > > After that enable gc logging and upload new information. > > > > > > Sent from my iPhone > > > > > > On Aug 26, 2013, at 6:45 PM, Ariel L [ariel2129@gmail.com](mailto:ariel2129@gmail.com) wrote: > > > > > > Hello, we are experimenting some problems with GC configuration of > > > Elasticsearch, > > > > > > we have elasticsearch 0.20.2 running in java 6.0.32 with 9 node with 32 > > > gb and 16 gb assigned to the elastic, in this cluster, we have one index > > > with 20 shard and 2 replicas, we have around 15k of request per minute. > > > > > > After a few hours, at least one node is expelled of the cluster and the > > > logs that we see are: > > > > > > [2013-08-26 20:33:36,030][INFO][monitor.jvm] [King > > > Bedlam] [gc][ConcurrentMarkSweep][16432][221] duration [18.9s], collections > > > [2]/[2.2m], total [18.9s]/[1.2m], memory [6.8gb]-\>[6.6gb]/[6.9gb], > > > all\_pools {[Code Cache] [5.9mb]-\>[5.9mb]/[48mb]}{[Par Eden Space] > > > [91.7mb]-\>[611.4kb]/[133.1mb]}{[Par Survivor Space] > > > [16.6mb]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.7gb]-\>[6.6gb]/[6.8gb]}{[CMS Perm > > > Gen] [42.6mb]-\>[42.2mb]/[82mb]} > > > > > > [2013-08-26 20:36:11,033][WARN][monitor.jvm] [King > > > Bedlam] [gc][ConcurrentMarkSweep][16439][223] duration [31.3s], collections > > > [2]/[2.4m], total [31.3s]/[1.7m], memory [6.8gb]-\>[6.7gb]/[6.9gb], > > > all\_pools {[Code Cache] [5.9mb]-\>[5.9mb]/[48mb]}{[Par Eden Space] > > > [68.8mb]-\>[32mb]/[133.1mb]}{[Par Survivor Space] > > > [16.6mb]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.7gb]-\>[6.7gb]/[6.8gb]}{[CMS Perm > > > Gen] [42.2mb]-\>[42.2mb]/[82mb]} > > > > > > [2013-08-26 20:38:54,467][WARN][monitor.jvm] [King > > > Bedlam] [gc][ConcurrentMarkSweep][16441][225] duration [37.8s], collections > > > [2]/[2.7m], total [37.8s]/[2.4m], memory [6.8gb]-\>[6.8gb]/[6.9gb], > > > all\_pools {[Code Cache] [6mb]-\>[6mb]/[48mb]}{[Par Eden Space] > > > [83.5mb]-\>[1mb]/[133.1mb]}{[Par Survivor Space] [0b]-\>[0b]/[16.6mb]}{[CMS > > > Old Gen] [6.7gb]-\>[6.8gb]/[6.8gb]}{[CMS Perm Gen] [42.2mb]-\>[42.2mb]/[82mb]} > > > > > > [2013-08-26 20:41:36,086][WARN][monitor.jvm] [King > > > Bedlam] [gc][ConcurrentMarkSweep][16442][227] duration [42.1s], collections > > > [2]/[2.6m], total [42.1s]/[3.1m], memory [6.8gb]-\>[6.8gb]/[6.9gb], > > > all\_pools {[Code Cache] [6mb]-\>[6mb]/[48mb]}{[Par Eden Space] > > > [1mb]-\>[9.8mb]/[133.1mb]}{[Par Survivor Space] [0b]-\>[0b]/[16.6mb]}{[CMS > > > Old Gen] [6.8gb]-\>[6.8gb]/[6.8gb]}{[CMS Perm Gen] [42.2mb]-\>[42.2mb]/[82mb]} > > > > > > [2013-08-26 20:44:23,710][WARN][monitor.jvm] [King > > > Bedlam] [gc][ConcurrentMarkSweep][16445][229] duration [40.1s], collections > > > [2]/[2.7m], total [40.1s]/[3.7m], memory [6.9gb]-\>[6.8gb]/[6.9gb], > > > all\_pools {[Code Cache] [6mb]-\>[6mb]/[48mb]}{[Par Eden Space] > > > [115.4mb]-\>[56.4mb]/[133.1mb]}{[Par Survivor Space] > > > [0b]-\>[0b]/[16.6mb]}{[CMS Old Gen] [6.8gb]-\>[6.8gb]/[6.8gb]}{[CMS Perm Gen] > > > [42.2mb]-\>[42.2mb]/[82mb]} > > > > > > I don't know how we can resolve this problem, we try diferents > > > configurations in the in the elasticsearch.in.sh but the problem > > > persist. > > > > > > would be great if someone could help with this. > > > > > > regards > > > > > > -- > > > 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). > > > 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). > > > For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). > > > > ## -- mvh > > > > Runar Myklebust > > > > -- > > 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). > > 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). > 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). For more options, visit [https://groups.google.com/groups/opt\_out](https://groups.google.com/groups/opt_out). ---
**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, 2:19am UTC](https://discuss.elastic.co/t/help-with-gc-configuration/13339/6 "2017-07-06T02:19:26Z")