load average goes to 80, 90 , 120 for java
load average goes to 80, 90 , 120 for java
Hi,
I have been running zimbra Release 8.6.0_GA_1153.RHEL7_64_20141215151110 RHEL7_64 FOSS edition for long time with 2200 users on CentOS Linux release 7.2.1511. But, today, load average increased suddenly. top command shows this.
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
4790 zimbra 20 0 21.829g 3.547g 22068 S 701.7 11.3 1151:20 java
this 4790 process is given below
ps aux |grep 4790
zimbra 4790 1115 11.3 22889732 3719096 ? Sl 17:43 1157:05 /opt/zimbra/java/bin/java -Dfile.encoding=UTF-8 -server -Dhttps.protocols=TLSv1,TLSv1.1,TLSv1.2 -Djdk.tls.client.protocols=TLSv1,TLSv1.1,TLSv1.2 -Djava.awt.headless=true -Dsun.net.inetaddr.ttl=60 -Dorg.apache.jasper.compiler.disablejsr199=true -XX:+UseConcMarkSweepGC -XX:PermSize=128m -XX:MaxPermSize=350m -XX:SoftRefLRUPolicyMSPerMB=1 -verbose:gc -XX:+PrintGCDetails -XX:+PrintGCDateStamps -XX:+PrintGCApplicationStoppedTime -XX:-OmitStackTraceInFastThrow -Xloggc:/opt/zimbra/log/gc.log -XX:-UseGCLogFileRotation -XX:NumberOfGCLogFiles=20 -XX:GCLogFileSize=4096K -Djava.net.preferIPv4Stack=true -Xss256k -Xms4096m -Xmx4096m -Xmn1638m -Djava.io.tmpdir=/opt/zimbra/mailboxd/work -Djava.library.path=/opt/zimbra/lib -Djava.endorsed.dirs=/opt/zimbra/mailboxd/common/endorsed -Dzimbra.config=/opt/zimbra/conf/localconfig.xml -Djetty.home=/opt/zimbra/mailboxd -DSTART=/opt/zimbra/mailboxd/etc/start.config -jar /opt/zimbra/mailboxd/start.jar --module=zimbra,server,servlet,servlets,jsp,jmx,resources,websocket,ext,plus,rewrite,monitor,continuation,webapp,setuid jetty.home=/opt/zimbra/mailboxd /opt/zimbra/mailboxd/etc/jetty.xml
sometimes load average goes to 80, 90 , 120 etc. it never happened before.
everything worked perfectly well until this morning.
here's bit of config settings
zmlocalconfig | grep java_heap
mailboxd_java_heap_new_size_percent = 20
mailboxd_java_heap_size = 6410
zmprov gs `zmhostname` zimbraHttpNumThreads
zimbraHttpNumThreads: 500
and
zmprov gs `zmhostname` zimbraLmtpNumThreads
zimbraLmtpNumThreads: 40
What has gone wrong? Hope to hear from you.
I have been running zimbra Release 8.6.0_GA_1153.RHEL7_64_20141215151110 RHEL7_64 FOSS edition for long time with 2200 users on CentOS Linux release 7.2.1511. But, today, load average increased suddenly. top command shows this.
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
4790 zimbra 20 0 21.829g 3.547g 22068 S 701.7 11.3 1151:20 java
this 4790 process is given below
ps aux |grep 4790
zimbra 4790 1115 11.3 22889732 3719096 ? Sl 17:43 1157:05 /opt/zimbra/java/bin/java -Dfile.encoding=UTF-8 -server -Dhttps.protocols=TLSv1,TLSv1.1,TLSv1.2 -Djdk.tls.client.protocols=TLSv1,TLSv1.1,TLSv1.2 -Djava.awt.headless=true -Dsun.net.inetaddr.ttl=60 -Dorg.apache.jasper.compiler.disablejsr199=true -XX:+UseConcMarkSweepGC -XX:PermSize=128m -XX:MaxPermSize=350m -XX:SoftRefLRUPolicyMSPerMB=1 -verbose:gc -XX:+PrintGCDetails -XX:+PrintGCDateStamps -XX:+PrintGCApplicationStoppedTime -XX:-OmitStackTraceInFastThrow -Xloggc:/opt/zimbra/log/gc.log -XX:-UseGCLogFileRotation -XX:NumberOfGCLogFiles=20 -XX:GCLogFileSize=4096K -Djava.net.preferIPv4Stack=true -Xss256k -Xms4096m -Xmx4096m -Xmn1638m -Djava.io.tmpdir=/opt/zimbra/mailboxd/work -Djava.library.path=/opt/zimbra/lib -Djava.endorsed.dirs=/opt/zimbra/mailboxd/common/endorsed -Dzimbra.config=/opt/zimbra/conf/localconfig.xml -Djetty.home=/opt/zimbra/mailboxd -DSTART=/opt/zimbra/mailboxd/etc/start.config -jar /opt/zimbra/mailboxd/start.jar --module=zimbra,server,servlet,servlets,jsp,jmx,resources,websocket,ext,plus,rewrite,monitor,continuation,webapp,setuid jetty.home=/opt/zimbra/mailboxd /opt/zimbra/mailboxd/etc/jetty.xml
sometimes load average goes to 80, 90 , 120 etc. it never happened before.
everything worked perfectly well until this morning.
here's bit of config settings
zmlocalconfig | grep java_heap
mailboxd_java_heap_new_size_percent = 20
mailboxd_java_heap_size = 6410
zmprov gs `zmhostname` zimbraHttpNumThreads
zimbraHttpNumThreads: 500
and
zmprov gs `zmhostname` zimbraLmtpNumThreads
zimbraLmtpNumThreads: 40
What has gone wrong? Hope to hear from you.
Re: load average goes to 80, 90 , 120 for java
-what does dmesg says?
-have you mounted some NFS shares? are they alive? df -h does works?
-have you mounted some NFS shares? are they alive? df -h does works?
Re: load average goes to 80, 90 , 120 for java
-have you mounted some NFS shares? are they alive? df -h does works?[/quote]
No NFS shares.. df -h works.
No NFS shares.. df -h works.
Re: load average goes to 80, 90 , 120 for java
my zimbra version is zimbra Release 8.6.0_GA_1153.RHEL7_64_20141215151110 RHEL7_64 FOSS editionindunil75 wrote:Hi,
What has gone wrong? Hope to hear from you.
This URl says ( https://www.zimbra.com/downloads/zimbra-collaboration/ )
Patch-10 has been issued on 10 May 2018 for 8.6.0 GA which includes fixes for reported bugs listed in the release notes.
I want to know if I Installed Patch-10, Does this patch 10 cover all the patches from patch 1 to patch 10 or Do I need to install separately all patches?
Hope to hear from you. is there patch to reduce load average and Cpu usage in zimbra Release 8.6.0? if available, pls update me since it's the most important thing..
Re: load average goes to 80, 90 , 120 for java
What about this patch? i.e - Patch 9
the below Url says
All Zimbra Collaboration 8.6.0 sites are recommended to install this patch. Patch 9 is cumulative with all the previous eight patches, so only Patch 9 is required in case you didn’t install the previous ones.
it has this below fix.
CPU load & latency when open mail with data:image/png:base64 inline image
https://blog.zimbra.com/2018/02/zimbra- ... 2017-8802/
Has anyone applied this patch?
the below Url says
All Zimbra Collaboration 8.6.0 sites are recommended to install this patch. Patch 9 is cumulative with all the previous eight patches, so only Patch 9 is required in case you didn’t install the previous ones.
it has this below fix.
CPU load & latency when open mail with data:image/png:base64 inline image
https://blog.zimbra.com/2018/02/zimbra- ... 2017-8802/
Has anyone applied this patch?
Re: load average goes to 80, 90 , 120 for java
Why don't you apply it after taking a suitable backup and find out if it solves your problem, after all it is a 'recommended' patch.indunil75 wrote:Has anyone applied this patch?
Re: load average goes to 80, 90 , 120 for java
Why don't you apply it after taking a suitable backup and find out if it solves your problem, after all it is a 'recommended' patch. [/quote]
Thanks for your inputs.
May I ask a few questions before applying the patch?
8.6 Patches can be found here
https://wiki.zimbra.com/wiki/Zimbra_Col ... _-_Patches
patch 9 url
https://files.zimbra.com/downloads/8.6. ... A_1224.tgz
patch 10 url
https://files.zimbra.com/downloads/8.6. ... A_1229.tgz
I think applying patch 10 is also OK since it covers patch 9.
Am I right? Can I go with patch 10?
Is my thinking process OK?
If I am right, I am going to take a back up and apply patch 10 since it covers patch 9.
hope to hear from you.
Thanks for your inputs.
May I ask a few questions before applying the patch?
8.6 Patches can be found here
https://wiki.zimbra.com/wiki/Zimbra_Col ... _-_Patches
patch 9 url
https://files.zimbra.com/downloads/8.6. ... A_1224.tgz
patch 10 url
https://files.zimbra.com/downloads/8.6. ... A_1229.tgz
I think applying patch 10 is also OK since it covers patch 9.
Am I right? Can I go with patch 10?
Is my thinking process OK?
If I am right, I am going to take a back up and apply patch 10 since it covers patch 9.
hope to hear from you.
Re: load average goes to 80, 90 , 120 for java
As all the patches are cumulative (it tells you that in the documentation) it would make sense to install the most recent patch.
Re: load average goes to 80, 90 , 120 for java
Sir, Thanks for your update.phoenix wrote:As all the patches are cumulative (it tells you that in the documentation) it would make sense to install the most recent patch.
I will apply patch 10 and UPDATE you. It's very likely to happen on Saturday.
Thanks for your wonderful support.
Have a nice day....
Re: load average goes to 80, 90 , 120 for java
mailbox.log shows thisphoenix wrote:Why don't you apply it after taking a suitable backup and find out if it solves your problem, after all it is a 'recommended' patch.indunil75 wrote:Has anyone applied this patch?
misc - Exceeded the max requests limit. Suspending org.eclipse.jetty.continuation.Servlet3Continuation@1b66838c
So, I ran below command
zmlocalconfig -e servlet_max_concurrent_http_requests_per_account=20
about 2 hours later, both load average and cpu usage decreased a lot.
Now, load average has gone to 8 , 9
cpu usage (%Cpu) has also gone down.
This is just an update..... Perhaps, It may help zimbra users.
Anyway, I am going to apply patch 10 ASAP.