[
https://issues.apache.org/jira/browse/HIVE-24650?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17652358#comment-17652358
]
hansonhe commented on HIVE-24650:
---------------------------------
I have the same problem on my production environment hive-3.1.2. Memory usage
is occupied high all the time,Even the service is stuck.
（1）jstat -gc \{hs2-pid}
S0C S1C S0U S1U EC EU OC OU MC MU
CCSC CCSU YGC YGCT FGC FGCT GCT
0.0 0.0 0.0 0.0 32768.0 0.0 67076096.0 67067900.3 202496.0
184741.9 0.0 0.0 7475 1686.253 142 5875.588 7561.841
(2) jmap -heap \{hs2-pid}
Heap Configuration:
MinHeapFreeRatio = 40
MaxHeapFreeRatio = 70
MaxHeapSize = 68719476736 (65536.0MB)
NewSize = 1363144 (1.2999954223632812MB)
MaxNewSize = 22900899840 (21840.0MB)
OldSize = 5452592 (5.1999969482421875MB)
NewRatio = 2
SurvivorRatio = 8
MetaspaceSize = 21807104 (20.796875MB)
CompressedClassSpaceSize = 1073741824 (1024.0MB)
MaxMetaspaceSize = 17592186044415 MB
G1HeapRegionSize = 16777216 (16.0MB)
Heap Usage:
G1 Heap:
regions = 4096
capacity = 68719476736 (65536.0MB)
used = 68669211960 (65488.06377410889MB)
free = 50264776 (47.93622589111328MB)
99.92685512406752% used
G1 Young Generation:
Eden Space:
regions = 0
capacity = 33554432 (32.0MB)
used = 0 (0.0MB)
free = 33554432 (32.0MB)
0.0% used
Survivor Space:
regions = 0
capacity = 0 (0.0MB)
used = 0 (0.0MB)
free = 0 (0.0MB)
0.0% used
G1 Old Generation:
regions = 2
capacity = 68685922304 (65504.0MB)
used = 68669211960 (65488.06377410889MB)
free = 16710344 (15.936225891113281MB)
99.97567136985357% used
> hiveserver2 memory usage is extremely high, GC unable to recycle
> ----------------------------------------------------------------
>
> Key: HIVE-24650
> URL: https://issues.apache.org/jira/browse/HIVE-24650
> Project: Hive
> Issue Type: Bug
> Components: HiveServer2
> Affects Versions: 3.1.0
> Environment: hive3.1.0
> Reporter: zhaojk
> Priority: Major
> Attachments: 1.png, 2.png, 3.png
>
>
> HDP's HiveServer2 is using 80GB of memory (HEAP is configured with 74GB), and
> when the memory is full, there will be frequent Full GC, and then the memory
> cannot be recycled, resulting in a service exception.Analyze memory usage.
> GC config:
> export HADOOP_OPTS="$HADOOP_OPTS
> -Xloggc:\{{hive_log_dir}}/hiveserver2-gc-%t.log -XX:ConcGCThreads=30
> -XX:ParallelGCThreads=30 -XX:+UseG1GC -XX:G1HeapRegionSize=8M
> -XX:+UseStringDeduplication -XX:MaxGCPauseMillis=1000
> -XX:InitiatingHeapOccupancyPercent=40 -XX:G1ReservePercent=15
> -XX:+PrintGCDetails -XX:+PrintGCTimeStamps -XX:+PrintGCCause
> -XX:+UseGCLogFileRotation -XX:NumberOfGCLogFiles=10 -XX:GCLogFileSize=100M
> -XX:+HeapDumpOnOutOfMemoryError
> -XX:HeapDumpPath=/home/hive/hs2_heapdump.hprof
> -Dhive.log.dir=\{{hive_log_dir}} -Dhive.log.file=hiveserver2.log
> The details at
> [https://blog.csdn.net/Small_codeing/article/details/112601226]
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
