One of the best places to start is to review the details in OEM itself. See
what the metrics tell you about performance. There may be additional
repository things that will help.
https://docs.oracle.com/en/enterprise-manager/cloud-control/enterprise-manager-cloud-control/13.3.1/emadm/overview-managing-manager.html#GUID-0CE3C2CE-A52F-4C8B-B5B1-88F427CC6792
<https://docs.oracle.com/en/enterprise-manager/cloud-control/enterprise-manager-cloud-control/13.3.1/emadm/overview-managing-manager.html#GUID-0CE3C2CE-A52F-4C8B-B5B1-88F427CC6792>
https://docs.oracle.com/en/enterprise-manager/cloud-control/enterprise-manager-cloud-control/13.3.1/emadm/management-servers.html#GUID-2891EEB1-9640-4409-B5E9-288975C5EF13
<https://docs.oracle.com/en/enterprise-manager/cloud-control/enterprise-manager-cloud-control/13.3.1/emadm/management-servers.html#GUID-2891EEB1-9640-4409-B5E9-288975C5EF13>
There’s also emdiag - EMDIAG Troubleshooting Kits Master Index (Doc ID 421053.1)
Both omsvfy and repvfy will help you learn about your env and performance.
Cleaning up agent communication, alert flooding , etc can also improve overall
OEM performance.
This blog is a bit old, but things remain the same. I would start here.
https://blogs.oracle.com/emmaa/oracle-enterprise-manager-12c-configuration-best-practices-part-1-of-3
<https://blogs.oracle.com/emmaa/oracle-enterprise-manager-12c-configuration-best-practices-part-1-of-3>
On Mar 28, 2020, at 10:50 AM, Herring, David <dmarc-noreply@xxxxxxxxxxxxx>
wrote:
Folks,
I'm looking for some help in terms of assessing performance of our weblogic
configuration as it relates to OEM. This past week we had been working with
Oracle on configuring backups to go to our new ZDLRA's and our screen
refreshes in OEM were painful. I have no issues with trying to assess the
performance of the OMR but the weblogic piece is where I'm lost. I did a bit
of searching on this and frequently the response, both from individual blogs
and Oracle doc, was something along the lines of "If you need to increase …
then use this command". Great, but I have no idea if I need to change xyz
parameter so I'm not going to change it unless I can show the given parameter
setting is an issue.
I'm looking for something where I can review components related to weblogic /
page refreshes for OEM and identify where bottlenecks are and attack those.
I believe all targets are discovered within OEM and they have performance
pages but again it just displays statistics over time and I have nothing to
compare against to say whether displayed items are good or bad.
Configuration details:
OEM 13.3
RHEL 6.10, 64 GB
Single OMS, single tier
Standard 2 domains for OEM, the admin and OMS
We have around 1,500 targets that are actively monitored w/ 125 agents
monitoring various envs, 65 jobs scheduled in OEM and around 15 DBA-level
concurrent connections at any point.
Thx in advance for any help.
Regards,
Dave
