David Jones

I've been having a look at this, and I so far haven't spotted how this could occur (a message going missing), particularly if the connection recovers later. That implies that...

I started off trying to recreate this issue with plain Kitura + Kitura-net, but haven't managed to reproduce either issue yet. Running in a constrained Docker environment and stressing a...

@bridger Apologies for the radio silence. I'm afraid I have been unable to recreate this issue locally - initially I thought I'd found something, but that turned out to be...

When you reproduced the issue, would I be correct in thinking that you have a Kitura WebSocket server running remotely, and are accessing it via a (local) iPad? If so,...

@kkieffer This looks and sounds reasonable, though I'm not a subject matter expert. @billabt any thoughts?

@ianpartridge The code for this change looks okay, what do you think?

The Travis failure on this job is unrelated to this PR, and has been happening on the CRON builds for master since about 3 weeks ago, eg: https://travis-ci.org/IBM-Swift/Kitura/jobs/648208574 Timeouts in...

@ianpartridge A reason to avoid validation during `init(from:)`: the TypeDecoder (used during Swagger generation) works by creating an instance using dummy values. Validation during init would cause this to fail,...

@ianpartridge and I discussed this. Something we'd like to be able to do is to represent validations in the OpenAPI (swagger) specification that Kitura can generate for a REST API....

There are some places where Kitura's configuration could be performed via Helm, but is currently hard-coded in the sample: - logging: hard-coded to use HeliumLogger at `.info` - level should...