We have a lot of test DataMiner systems, which we currently all monitor with Microsoft Platform driver. However, this only allows us to gather information about the server (CPU, memory, disk,...). To a certain degree, we can monitor the different SL* processes using the task manager, but this does not allow us to monitor the state of the DataMiner. For example, amount of elements, protocols, services, redundancy groups, alarms,...
We can see all this information on every individual DataMiner agent, but is there a way we can "poll" this information into a central DataMiner so we can do aggregated trending/alarming on this information without having to write custom scripts/tools?
Hi Jens,
In the past we did something similar by using the Skyline DMP Replication connector.
We created an element with this connector on all of the DMP's (or DMA's).
On one central DMA you can then re-create the same replicated elements for these external agents or probes. (How to setup a replicated element)
That way you can monitor the health of multiple agents from one location.
Via these elements you can see a lot of information about the DataMiner platform you are monitoring. Some examples are: Alarm counts of all severities, the DM software version, Server processor load & Physical mem usage. Also the Number of elements and active elements is visible.
There is also a feature to perform DM restarts or even Server reboots via these replicated elements.
On the extended pages you can then also enable lists such as (Elements, Views, Services, Alarms). These lists support filtering, so when you are only interested in a subset of data, you can easily narrow it down.
If you have more questions on this, please ask.
If your system is connected to CDMR, watchdog reports are forwarded to the CDMR server.
In those reports, there are some statistics about the DataMiner, those reports are parsed and visualized by the CDMR agent driver.
More info about CDMR can be found here