Agent status as a time series data

Newbie question. I am exploring elastic for our needs. We have a lot of filemonitor agents. these agents can go down sometimes and we want to monitor this via Kibana

my understanding, i need to save the status of the agent in a database/mongo etc., then continuously publish this data to elastic search in intervals of say 5 minutes or whatever, that way this becomes a timeseries data rather than just state data. Then when agent goes down and the database will have another value and that get published until status changes again in database.

is this the approach i should take to monitor state information. is there a better way. Is elk a good solution for this?

The Elastic Stack is a great solution to this, yes.

thx. Is there a easy way to keep pumping same data to elasticsearch until a change happens?

Where is the data being saved to now?

This is a POC i am doing. Planning to save it mongodb and then pump it to elasticsearch

Why not just send it straight to Elasticsearch?

i subscribe to agent events, those events can be many things. running no running is just one of the events. i was planning to store them in different collections in mongo and then feed time series version of them into elastic..

for example, number of files being monitored comes as an event message which i want to show as a graph as well. hoping i am makingset sense.

my question is how can i send same data over and over until state changes..is there an option in elastic to auto create previous value until new value comes in sort of config perhaps?

For Elasticsearch you would just send each event as an individual time based event. Then go from there.

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.