jdef

I've thought more about this and I'm still not sold. This smells like a bigger infra problem, not a volume one. E.g. we could intro a controller rpc called "NodeGone"...

> > I've thought more about this and I'm still not sold. This smells like a bigger infra problem, not a volume one. E.g. we could intro a controller rpc...

Out of order CSI calls are not spec compliant. I've said as much in the KEP On Tue, Dec 28, 2021, 3:06 AM Yuiko Mouri ***@***.***> wrote: > This comment...

> ... protobuf complains. Do you mean the CO or plugin is complaining (because they're actually validating the protobuf fields according to the CSI spec, as they should be)? The...

I think we should probably amend the spec and extend the length limit for path fields. Validation implementations could then follow suit. On Thu, Feb 21, 2019 at 11:42 PM...

A quick Google search suggests that 4k is the length limit for common Linux filesystems. I guess Windows systems used to have a 255? 260? limit, but apparently that's overridable...

Is this blocking you? On Thu, Mar 25, 2021, 10:31 AM JeeShao ***@***.***> wrote: > Why is there no dependency management like gomod in this repo? > > — >...

It's a bit primitive, and not idiomatic, but some versions *are* pinned here: https://github.com/container-storage-interface/spec/blob/master/lib/go/Makefile On Mon, Mar 29, 2021 at 3:16 AM JeeShao ***@***.***> wrote: > If open-source software is...

Thanks for providing additional context @msau42. It's still not clear to me how including the "total" capacity helps resolve these things. If the CO can't reason about "available" capacity due...

Thanks for the example, your use case is much more clear now. I think using total_capacity in the way that you want will break down for cases where storage backends...