NdibeRaymond
commented
2 years ago I am taking my time to think through this given how broad-reaching it can potentially be.
- We need a monitoring system advanced enough to monitor docker both in swarm mode and normal mode (this is to allow us monitor any deployment of our codebase whether in swarm mode not).
- We also need to seriously consider building a monitoring system that can be used to monitor not just zubhub.unstructured.studio but also agastya.zubhub.co and any other deployment.
- The monitoring system probably needs to also be capable of monitoring vm health (or nodes in the case of a swarm depoyment), docker daemon health and any other thing that needs monitoring.
I am not 100% sure of how this can be done but I'm currently researching and thinking through the problem.
right now we have very limited monitoring system. Often relying on the basic traffic monitoring provided by digitalocean. We have no application specific monitoring even though a couple of the constituent parts have been implemented