We expect to see as per doc https://docs.appdynamics.com/display/PRO44/Server+Network+Metrics
can you share screenshots depicting the behaviour in controller UI if required we will ask machine aegnt node logs.
... View more
For users who have reviewed the Confirming Backend Discovery and Instrumentation article, some JDBC database or remote service database calls may be missing from the Controller UI
One potential cause could be when an Agent detects an API exit call/backend class and method that is supported out of the box as part of instrumentation, but the parent API is part of discovered Business Transactions (BTs).
class4.method4 is an exit call.
preparedStatement.executeQuery() Class1.method1 => Class2.method2 => com.userapp.Class4.method4 => Class3.method3 => Class5.method5 => Class6.method6
As a solution, a user can define custom POJO rule on the parent class API (in the above example, class2.method2 or class1.method1).
However, if the user does not know the details of the parent API stack (in the above example, no details on class2.method2 or class1.method1) use the following node property:
name: debug-interceptors type: string value: com.userapp.Class4/method4 Note: A dd to only one node instead of tier if not using a cluster environment.
This node property will return the reverse code stack in the Agent logs once the load is made with the node property in place.
The next step would be to share these Agent logs with the support team who can suggest the right POJO entry point based on the stack dump output.
Note: The example node property will result in more logging in the Agent logs; therefore, the node property should be removed from the UI after about 5 minutes of load.
... View more
The following FAQ is specific to non-High Availability Controller environments.
For production Controllers that will not start, contact Support and include the Controller logs located in the installation directory.
I completed the Controller installation, but the browser response shows a 404 error. How to I resolve this?
Usually, this issue is due to a typo in the Controller URL.
I can not reach the Controller URL from other machine using the host name/DNS provided during controller installation. The installation was a success, and messages in the <Controller_install_dir>/logs/startAs.log file indicate that the domain started just fine. How do I proceed?
First, check if the host name used for the Controller installation is mapped in the Controller host's file.
In Linux machines, this is located at /etc/hosts and in Windows machines, it is located at C:\Windows\System32\Drivers\etc\hosts. It's generally a good idea to take a backup before making any changes. Next, if the mapping entry exists in the host file, check if there are any firewall or proxy rules that are blocking access to the host.
Why does a 500 error appear in the browser after restarting the Controller?
This may be the same issue as outlined here: Why does the Controller take longer to start when App Server and MySQL are restarted?
Why are there no logs in the <Controller_install_dir>/logs/server.log file, apart from a JVM arguments list, and the following error message: "There is a process already using the admin port 4849"?
First, ensure that no other processes are using port 4848 with the command netstat -a | grep 4848 in the machine's command line.
Then check if the Controller host name entry exists in the operating system host file. In Linux machines this is located in the /etc/hosts directory. Take a backup before making any changes.
The operating system restarted abruptly, and now the Controller UI is not loading. How do I resolve this?
Plan ahead to avoid sudden operating system shutdowns by adding a Controller Stop script to the operating system shutdown process. This will stop the Controller before the OS goes now, which will ensure proper Java and MySQL shutdowns.
Linux: Execute a command before shutdown Windows: Run a script just before shutdown or reboot on Windows Home edition
It is also recommended that users validate if the Java processes and MySQL have both shutdown. Linux users should run commands ps -ef | grep java or ps -ef | grep mysql . Windows users should check Task Manager. Do not kill any processes, and contact support if there is only one process for Java or MySQL.
If the processes are running, just wait. If both processes are not running, start the Controller MySQL process, then start the UI process. Windows users can do this via Task Manager.
Linux users, run the following commands:
shell> cd <Controller_install_dir>/bin shell> ./controller.sh stop-appserver shell> ./controller.sh stop-db shell> ./controller.sh start-db shell> ./controller.sh start-appserver
If the UI still does not start, contact Support with the Controller log files.
Why can't I reach the Controller port? I see that the server.log files have activity, the Controller-related processes are running, and I can ping the Controller host, but the port is not reachable.
Check the server.log files for the following message:
Network listener http-listener-1 on port 8090 disabled per domain.xml
If the port was disabled on a recent restart, first take a backup of the domain.xml file (located at <Controller_install_dir>/appserver/glassfish/domains/domain1/config/) and run the following commands:
/opt/AppDynamics/Controller/bin/controller.sh enable-http-listeners /opt/AppDynamics/Controller/bin/controller.sh stop-appserver /opt/AppDynamics/Controller/bin/controller.sh start-appserver
... View more
Go through below steps in sequence to get more details on how machine agent registration and association works:
1) How machine agent association works, in brief based on uniqueHostId (default or specifically passed)
2) Provide only host name in recent machine agent version if usins <sim-enabled> set to true (default setting) if using older machine agent version, sim setting does not exists, and if you want to associate to particular app server node, make sure default or value for -Dappdynamics.agent.uniqueHostId= arg value for both app server and machine agent node startup script is same
3) Starting machine agent is like any java process, shell> java -jar <MachineAgent_install_dir>/machineagent.jar
Hope that information helps. Refer doc which are more detailed https://docs.appdynamics.com/display/PRO43/Install+the+Standalone+Machine+Agent
... View more
I do not see other option as of now for scenario where machine agent itself is down, however can you go through below feature doc https://docs.appdynamics.com/display/PRO43/Add+Tasks+for+Workflow+Steps
... View more
i have sent login details for new user in private message, login UI using that and create new user with all permisisons that could create /update all other users and delete user i created post relogin with new user you created hope that resolves
... View more