We have updated our Terms of Service, Code of Conduct, and Addendum.

Trying to set up monitoring for expanding health checks on various endpoints.

Options
mferris
mferris Posts: 27 admin
in General Discussions

Hey everyone! General question to see if I'm missing something. We're about to configure a Health check that is going to be ran across dozens of endpoints and will inevitably expand in the future even more. Looking to make sure this is as painless as possible so I'm trying to figure out the "best practice" for doing this. We're essentially monitoring overall health pages exposed by some of our containerized applications. All the urls will be formatted the same (<environment>.http://site.com/status/json.php|site.com/status/json.php). Ideally we'd have a way for this to dynamically update based on some outside source, second best would be for us to maintain a lookup on cribl of all the environments. Just looking for a way that I don't need to manually create sources and rely on my dev ops teams to tell me when new environments are spun up. Any thoughts?

Answers

  • pdugas
    pdugas Posts: 14 mod
    Options

    May want to look at Collectors. Code a "discover" task to build the list of URLs then run a "collect" task for each one.

  • pdugas
    pdugas Posts: 14 mod
    Options

    Is this in Kubernetes?

  • Brendan Dalpe
    Brendan Dalpe Posts: 201 mod
    edited October 2023
    Options

    what kind of envs?

    If it's k8s, you can have them inject the Helm chart at the cluster level which and use the Fleet Mapping rules to put them into the right group with a "default" catch-all that gets monitored for new nodes. Anything there triggers an alert to create an appropriate fleet.

  • mferris
    mferris Posts: 27 admin
    Options

    We're deploying it through stream current since it seems edge fleets in k8s will be quite expensive for us to maintain. This is just going to be a REST call. These environments do represent the same service hosted on multiple different clusters in k8s though. To the first comment, I've never looked into discover tasks, so I'll work to learn more about those and see how to actually apply them to these situations!

  • Brendan Dalpe
    Brendan Dalpe Posts: 201 mod
    Options

    > edge fleets in k8s will be quite expensive for us to maintain Would love to know more about this statement

  • mferris
    mferris Posts: 27 admin
    Options

    I've already bubbled it up through Dan/John a little bit. Can definitely hop on a chat sometime to discuss the concerns with you though if you want to get some time together

Categories