Zimbra Forums

This appears to be working now. I increased the number of Imap and Http threads and seems to have resolved the issue. Thank you everyone!

I have the same symptoms

after migration from version 8.7 to 8.8.5 the server started to get extremely slow

java is consuming all CPU

To solve, I have to stop the services and restart but the issue comes back after 1/2 days

ubuntu server 14.04
16GB RAM
VM with 8 VCORE

rui wrote:I have the same symptoms

after migration from version 8.7 to 8.8.5 the server started to get extremely slow

java is consuming all CPU

To solve, I have to stop the services and restart but the issue comes back after 1/2 days

ubuntu server 14.04
16GB RAM
VM with 8 VCORE

You're going to have to give a bit more information that 'java is consuming all CPU'. Too many CPUs can also cause problems and I wouldn't have thought you'd need eight of them and more doesn't increase performance, please read this article: https://wiki.zimbra.com/wiki/Performanc ... re_vSphere

How about some further information about the exact problem, which threads have you read on this topic and have you tried any of the solutions? What type of workload (imap or web ui) and how many users are on the server?

Every time you post should give the full output of the following command (so we know the exact version & o/s you're using):

zimbra@mail:~$ zmcontrol -v
Release 8.8.5.GA.1894.UBUNTU14.64 UBUNTU14_64 FOSS edition.

workload on imap and web ui

590 Users Account

Try:

zmprov ms `zmhostname` zimbraImapNumThreads 500
zmprov ms `zmhostname` zimbraHttpNumThreads 600
zmprov mcf zimbraLmtpNumThreads 40


TOP:
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
40767 zimbra 20 0 9.831g 5.130g 15708 S 100.7 32.7 261:50.52 java
42143 zimbra 20 0 56480 11220 3152 S 1.7 0.1 0:43.29 zmlogger
10435 root 20 0 25220 1780 1100 S 0.7 0.0 28:29.51 top
41937 zimbra 20 0 22380 1856 1096 R 0.7 0.0 0:00.07 top
42409 zimbra 20 0 45928 8088 2380 S 0.7 0.0 0:59.55 zmstat-proc
7 root 20 0 0 0 0 S 0.3 0.0 6:22.19 rcu_sched
14 root 20 0 0 0 0 S 0.3 0.0 0:58.28 rcuos/6
39068 zimbra 20 0 80.382g 149284 15212 S 0.3 0.9 6:05.59 slapd
40483 syslog 20 0 190444 1152 708 S 0.3 0.0 0:11.42 rsyslogd

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
40767 zimbra 20 0 9.837g 5.131g 15712 S 101.0 32.8 269:18.97 /opt/zimbra/common/bin/java -Dfile.encoding=UTF-8 -server -Djava.awt.headless=true -Dsun.net.inetaddr.ttl=60 -Dorg.apache.jasper.compiler.disablejsr199=true -XX:+UseCo+
42004 zimbra 20 0 859416 17812 2724 S 1.3 0.1 0:43.71 /opt/zimbra/common/bin/httpd -k start -f /opt/zimbra/conf/httpd.conf

zimbra@mail:~$ free -m
total used free shared buffers cached
Mem: 16042 15428 614 0 384 5129
-/+ buffers/cache: 9914 6128
Swap: 47087 289 46798

CLUSTER HYPER-V

verticon wrote:This appears to be working now. I increased the number of Imap and Http threads and seems to have resolved the issue. Thank you everyone!

Can you elaborate on what the values you set?

Our server is suddenly slowing down to a crawl for web users after 24-48 hours since upgrading from 8.7.11 to 8.8.6

This is what we have:

Hi guys,
Thank you for reaching out and identifying the issue. We're on it! As we conduct a high priority investigation to determine if or what patches might be necessary, check back here for updates in the next few of days. Thanks again!

Here is the same Problem since Upgrade to 8.8.6
After about 24-48 hours the java-Process consumes 100% CPU. 55 Users in a Multidomain Environment. Eight Domains.
VMWare with four CPUs and 12GB RAM.