Observability Pack for static websites

[zstix]

Summary

Adds some basic configuration for a static site, such as one created with Gatsbyjs. This pack includes:

• A synthetics monitor that will ping the website (with retries)
• Alert conditions for high JS errors and low apdex
• A troubleshooting dashboard

I took some guesses as to how we would handle variables. This was my first attempt at making a pack, I'd love some feedback!

[jpvajda]

adding to in review column so we can get this live. 🥳

[polfliet]

Great contribution Zack.
I had a few small comments and then it would be awesome to merge this pack in, thanks!

[polfliet]

Hi @bpeckNR! You have probably heard that we are moving to this new repository for observability packs (previously quickstarts). Here is a contribution from Zack that includes a Synthetic check and I would be great to get your review since you are our Synthetics expert! Thanks a lot!!

[zstix]

@polfliet I've made the requested updates. Let me know if you need anything else!

[polfliet]

thanks @zstix !

[zstix]

Unfortunately, I don't think the functionality needed by this work is ready yet. More specifically, we are running into the following issues:

• Browser alert condition types are not supported
• NRQL field is required on alert conditions (not relevant for browser alert conditions)
• No way to reference the static site's URL to ping (variable support)

For those reasons, this pack doesn't provide sufficient value and I'm going to close this PR. I just spun up a ticket to track the feature request for variable support (#52).

[bpeckNR]

Hi @bpeckNR! You have probably heard that we are moving to this new repository for observability packs (previously quickstarts). Here is a contribution from Zack that includes a Synthetic check and I would be great to get your review since you are our Synthetics expert! Thanks a lot!!

I know this PR has been closed, but I wanted to circle back on the review that was requested. It looks like the sample Synthetic monitor included a few retries but would have needed to be a scripted API monitor.

For a static site like this one, it may make sense to just use the built in Synthetics ping monitor type. While by default that uses a HEAD request, it can be configured to use a GET request and even validate the response.

As far as the retries built into the script, please note that the Synthetic environment includes 2 built in retries before a job shows up as failed.

My suggestion is to simplify this and just use a ping type OOTB Synthetics monitor.