Monitoring dynaTrace Server Availabillity
Hi there
We have several dynaTrace Servers. When a dynaTrace server goes down, the alert triggered by the Self-Monitoring System Profile is only sent out when the server has been restarted and is functioning again. We can monitor a dynaTrace server with another dynaTrace server to make sure that the alert about the outage goes out when the server stops monitoring (becomes unavailable). We have a Host Agent on the dynaTrace server, so we know that the OS is still functioning, but the dT server is might have stopped.
What would be the best way to alert on that the dT server is not available? Even with a port check the server port like 2021) might still be open but it is not monitoring anymore. Would a URL monitor to port 8020 be an option?
Thanks
Hi Anton,
Have you tried the following monitors?
Windows Service Status Monitor Plugin
Windows Process Monitor Plugin
Thanks
Josh
Hi Josh
All of the dynaTrace servers are Linux based. Unfortunately we cannot SSH directly to a server and due to security reasons have to go via a security gateway on another machine to get access to the machine. We also are not allowed to put a collector on this security machine.
Anton
Hey Anton,
Sounds like a challenge!
It looks like the only way we can monitor it then is externally via URL monitor or TCP?
Josh
Thanks Josh
This is what I feared.
Anton
Hi Anton
I understand that you want to get a notification if the dT Server is not working as expected. But - did you know that we have a built-in mechanism that would restart the server in case its not working properly any longer? WE call it the Watch Dog. Its there to restart the server in case there are problems.
If you stilll want to monitor the dT Server from the outside using your own tools I suggest you go ahead and call one of the REST Interfaces, .e.g the Server Version REST call: Server Management (REST)
Hi Andreas
I know about the Watchdog. The problem is that the server sometimes gets so busy that monitoring and alerting stops for a long time. The Watchdog still thinks that the server is running although it is not. I think I might use the Web Transaction Monitor Plugin to log in to the Dashboards.
Anton
Anton - can you open a support ticket for this? If you really have the problem that the server becomes unresponsive and not even the self-monitoring incidents are triggered then the engineering team should take a look at this
Hi Andreas,
Thanks! I have already. Just trying to manage it until we get a resolution.
Anton
