Hello friends,
I am monitoring application which has different components. I would like to monitor dependencies between different components or different servers (if each component is deployed on different server).
So e.g. Sever1 --> Server2 --> server3 and server4.
If server1 goes RED, its downstream servers (2,3 and 4) should show RED.
if Server3 goes RED, all other should not go RED.
I find it hard to give any kind of recommendation based on those vague information. What is the shape of your monitoring data and what does "going RED" mean? And what kind of output do you expect? A table with red rows? A graph with red bars? A notification mail?
I am currently monitoring servers using Nagios.
I am taking logs from Nagios in to Elasticsearch. At the moment, I can see the status as Hostname , it's parent and it's status in Metric Visualisation.
e.g.
HOST PARENT STATUS
Server1 ROOT UP
Server2 Server1 DOWN
Server3 server1 UP
Server4 Server2 UP (Ideally it should show Down).
I think there is no way of formatting a field depending on the data in another field or even row. In order to achieve that you could already include the information about the dependency status when ingesting the data from Nagios. Maybe your ingestion tool can add it?
I can't think of a built-in way to visualize a hierarchy. You might be able to generate a textual hierarchy using some mustache templating in this community plugin: https://github.com/PhaedrusTheGreek/transform_vis
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.