topic Not getting Containers Data in Splunk AppDynamics

Not getting Containers Data

CommunityUser — Tue, 19 Sep 2017 09:27:42 GMT

Hi,

I followed this article related to complete visibility for Docker containers. But not getting data on Server-->Containers .

https://blog.appdynamics.com/product/complete-visibility-into-docker-environment-with-appdynamics/

Passing the following argument while brining up the machne agent :

-Dappdynamics.agent.reuse.nodeName=true -Dappdynamics.agent.reuse.nodeName.prefix=dockerreport

Forgot to mention that the dashboard created dynamically from extension is not showing data as well.

Also, with every sampling duration , a new dashboard is getting created for both TCP and Unix sockets, which is kind of annoying.

Using the latest release of machine agent.

As I am new to AppDynamics, some of the articles explaining the solution are restricted.

Attaching the log file and Docker monitoring YAML file for reference.

Screen capture of Docker Monitoring extension with multiple dashboards being created.

Thanks in advance for the help.

Thanks

Milind

Re: Not getting Containers Data

CommunityUser — Tue, 19 Sep 2017 10:19:51 GMT

As I could not edit the original post, hence a reply...

Access to the following community page is denied and it's published on the Appd Blog site.

http://community.appdynamics.com/t5/eXchange-Community-AppDynamics/Docker-Monitoring-Extension/idi-p/14749

Is it possible to get access to the content.

Thanks

Milind

Re: Not getting Containers Data

Atyuha_Pal — Wed, 20 Sep 2017 06:39:39 GMT

Hi Milind,

I can see the below error in the log

i) [AD Thread Pool-Global0] 19 Sep 2017 09:12:27,158 WARN SystemAgentPollingForUpdate - Invalid response for configuration request from controller/could not connect. Msg: Fatal transport error while connecting to URL [/controller/instance/195756/systemagentpolling]

Could you please check the connectivity between agent and controller.

ERROR ProcessExecutor$ErrorReader - Process Error - sudo: unable to resolve host ip-10-0-1-64

ii) Also, with every sampling duration , a new dashboard is getting created for both TCP and Unix sockets, which is kind of annoying.

It will create a dashboard for each host. This is expected.

iii) metricPrefix is mandatory

please pass either unixSocket or tcpSocket based on what is enabled in configure.yaml file

iv) The Docker metrics will be available under "Custom Metrics / Docker" under Metric Browser#Application infrastructure performance.

v) The link you have reffered could be old one. Please refer to the below link

https://www.appdynamics.com/community/exchange/extension/docker-monitoring-extension/

Thanks,

Atyuha

Re: Not getting Containers Data

CommunityUser — Wed, 20 Sep 2017 10:54:18 GMT

Thanks Atyuha for the response.

My Docker host (unable to resolve host ip-10-0-1-64)is an AWS instance and it will not expose this IP . Is there anything required to resolve this problem?

ii) Also, with every sampling duration , a new dashboard is getting created for both TCP and Unix sockets, which is kind of annoying.

It will create a dashboard for each host. This is expected.

There is only one host , so not sure why it's creating everytime the sampling data is sent from agent to controller. Can you please share your view?

iii) metricPrefix is mandatory

please pass either unixSocket or tcpSocket based on what is enabled in configure.yaml file

Believe , this is enabled and getting the JSON data on tcpSocket. Am I doing anything wrong? Please suggest.

iv) The Docker metrics will be available under "Custom Metrics / Docker" under Metric Browser#Application infrastructure performance.

I can cofirm that the metrics are visible under Metrics Browser for the containers and Host machine.

Can you share a sample config.yml file for the extension.

Thanks again for the responses and help.

Milind