W. Trevor King

> Kubernetes has PostStart/PreStop container lifecycle hooks... The spec doesn't have pre-stop hooks, because it doesn't have a `stop` operation. It has a [`kill`][1] operation, which you can use for...

On Mon, Oct 02, 2017 at 09:16:14AM +0000, Adam Medziński wrote: > To be honest, as you go along this reasoning, no hook is guaranteed > to be run because...

> Should I make a pull request with proposed changes or more opinions is needed? Working up a pull request is more work for you, and the maintainers may end...

There was some discussion of this in today's meeting, although it looks like [collabot decided to check out][1] and miss the fun :p. Here's my local log: 14:07 @wking I'd...

On Wed, Jan 10, 2018 at 11:23:04PM +0000, Aleksa Sarai wrote: > The GPU usecases in particular show how useful hooks are (they even > modified the hook design in...

On Sat, Nov 05, 2016 at 06:24:49PM -0700, Aleksa Sarai wrote: > … we should consider adding some OCI struct tags that add some > information about the OCI specification...

Currently runtime-tools is [using `omitempty`](https://github.com/opencontainers/runtime-tools/blob/553c260f75393e07530883481397cbfbc36b2aae/validate/validate.go#L557-L627) for this purpose.

This is moving *against* #680 and #813. Things like “Linux runtime $X ignores Windows property $Y” are acceptable, because the runtime-caller should know they are calling a Linux runtime and...

> I don't think a VM is a platform. Which is why I prefer "protocol", but [the maintainers wanted "platform"][1]. > ... you could potentially have a vm platform and...

On Thu, Mar 09, 2017 at 08:01:24AM -0800, Vincent Batts wrote: > @wking I read this, and am not real clear on the point? I think we need to have...