Since this has transitioned to using an in repo helm chart, the helm chart documentation should be checked, and documented. The current helm chart in this repo, seems to be built from several layers of templating which makes it harder to understand, and the correctness of this documentation all the more essential.

• The last Helm Stable repo version had the appropriate section documenting the

Sorry, this is a verbal report from a customer, I don't have any logs or repro steps just yet.

According to customer S, they see failures in the join process due to some sort of timeout, if ICMP isn't enabled between hosts. This was reported to be related to a ping, however, based on the discussion I suspect it's more likely when the join process tests the endpoints for whether it's joining a

Web metrics currently support only GET requests against a URL, but I can imagine there are use cases where a different request method (i.e. POST, PUT) may be desired instead, along with parameterized payloads to go along with it.

The Serverless Framework allows functions to be packaged into a directory, stored, and deployed in a separate step. This aligns well with CI/CD workflows, including the ones implemented in vili.

vili should support reading a re

I'd be nice to have a section describing how to drain tasks using the recently introduced kill policy.

See mesosphere/marathon#712 (comment) and https://issues.apache.org/jira/browse/MESOS-4547

The screenshot shown here https://cicd.serverlessworkshops.io/buildpipe/done.html needs to be updated to show the correct name to match the pipeline. ExecuteChangeSet is now Deploy.

See:

• https://apiblueprint.org/
• https://github.com/danielgtaylor/aglio

API Blueprint is basically markdown for API definitions. There's a ton of great tooling for it which even includes mock servers. Aglio is a parser/renderer written in nodejs, so at least from that standpoint the tooling is better than slate which uses ruby.

See some examples:

• [default](http://htmlpreview.github.io

We should modify our qa-script to allow Mac users (or just for @carkod xD) to run our deployments locally using Multipass.

One option could be to pass the konf.py output to kubectl but for some reason, the following doesn't work:

./konf.py production sites/canonical.com.yaml | multipass exec ubuntu-lts -- microk8s.kubectl apply -f -

Strangely the following does work as expected:

Make sure to update all official plugins documentation to fbl page.