-
Bug
-
Resolution: Duplicate
-
P3
-
None
-
9
-
x86_64
-
linux_ubuntu
FULL PRODUCT VERSION :
java version "9-ea"
Java(TM) SE Runtime Environment (build 9-ea+131)
Java HotSpot(TM) 64-Bit Server VM (build 9-ea+131, mixed mode)
FULL OS VERSION :
Linux 3.13.0-92-generic #139-Ubuntu SMP Tue Jun 28 20:42:26 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux
A DESCRIPTION OF THE PROBLEM :
We have ran our application on JDK 9-ea+131 and generated workload similar to one we have on production. We have noticed unacceptably long mixed GC pauses - more than 3 seconds long. Actually all mixed GC pauses where much longer than we have on JDK 1.8.0_102 in exactly similar conditions.
We have used following flags with JDK 9:
-XX:+DisableExplicitGC -XX:+PerfDisableSharedMem -XX:+UnlockDiagnosticVMOptions -XX:+PrintSafepointStatistics -XX:+UseG1GC -XX:MaxGCPauseMillis=40 -Xms64g -Xmx64g
It was found thatJDK-8136677 causes this since disabling Adaptive sizing for IHOP resolve the issue.
Attaching a part of debug GC log.
Reproducible on 9-ea+135
THE PROBLEM WAS REPRODUCIBLE WITH -Xint FLAG: Did not try
THE PROBLEM WAS REPRODUCIBLE WITH -server FLAG: Did not try
REGRESSION. Last worked in version 8u102
ERROR MESSAGES/STACK TRACES THAT OCCUR :
Part of debug GC log.
https://www.dropbox.com/s/ahmbutoai2j3ehh/gc_log_07_09_16.log.zip?dl=0
Long pauses can be found near 44320, 48900, 54470, 64900 seconds
REPRODUCIBILITY :
This bug can be reproduced always.
CUSTOMER SUBMITTED WORKAROUND :
-XX:-G1UseAdaptiveIHOP brings results on par with JDK8
java version "9-ea"
Java(TM) SE Runtime Environment (build 9-ea+131)
Java HotSpot(TM) 64-Bit Server VM (build 9-ea+131, mixed mode)
FULL OS VERSION :
Linux 3.13.0-92-generic #139-Ubuntu SMP Tue Jun 28 20:42:26 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux
A DESCRIPTION OF THE PROBLEM :
We have ran our application on JDK 9-ea+131 and generated workload similar to one we have on production. We have noticed unacceptably long mixed GC pauses - more than 3 seconds long. Actually all mixed GC pauses where much longer than we have on JDK 1.8.0_102 in exactly similar conditions.
We have used following flags with JDK 9:
-XX:+DisableExplicitGC -XX:+PerfDisableSharedMem -XX:+UnlockDiagnosticVMOptions -XX:+PrintSafepointStatistics -XX:+UseG1GC -XX:MaxGCPauseMillis=40 -Xms64g -Xmx64g
It was found that
Attaching a part of debug GC log.
Reproducible on 9-ea+135
THE PROBLEM WAS REPRODUCIBLE WITH -Xint FLAG: Did not try
THE PROBLEM WAS REPRODUCIBLE WITH -server FLAG: Did not try
REGRESSION. Last worked in version 8u102
ERROR MESSAGES/STACK TRACES THAT OCCUR :
Part of debug GC log.
https://www.dropbox.com/s/ahmbutoai2j3ehh/gc_log_07_09_16.log.zip?dl=0
Long pauses can be found near 44320, 48900, 54470, 64900 seconds
REPRODUCIBILITY :
This bug can be reproduced always.
CUSTOMER SUBMITTED WORKAROUND :
-XX:-G1UseAdaptiveIHOP brings results on par with JDK8
- duplicates
-
JDK-8159697 Adaptive G1HeapWastePercent, G1MixedGCLiveThresholdPercent, G1OldCSetRegionThresholdPercent, and G1MixedGCLiveThresholdPercent and G1MixedGCCountTarget
- Open
- relates to
-
JDK-8043573 G1NewSizePercent is too coarse on very large heaps
- Open
-
JDK-8136677 Adaptive sizing for IHOP in G1
- Resolved
-
JDK-8162928 Micro-optimizations in scanning the remembered sets
- Resolved
-
JDK-7146242 G1: alternative heuristic on deciding which old regions not to collect
- Closed