Implementing DeAllocate()

[Ed Warnicke]

The Issue for DeAllocate() seems to have stalled.   Would folks be OK with NSM folks pushing a PR for a simple implementation?

Or is there another recommended way to reclaim and recycle a HW resource once the Pod it was Allocated for has stopped and thus is definitionally

not using it anymore?

I've added it to the Jun 20 agenda in case there is desire for discussion.

Ed

[Vishnu Kannan]

What is your use case? 

--
You received this message because you are subscribed to the Google Groups "kubernetes-wg-resource-management" group.
To unsubscribe from this group and stop receiving emails from it, send an email to kubernetes-wg-resource...@googlegroups.com.
To post to this group, send email to kubernetes-wg-re...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/kubernetes-wg-resource-management/0a417973-235c-445e-85d9-4fb3e1ab5ef9%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Ed Warnicke]

Our use case is pretty straightforward.  A piece of HW has been allocated to a Pod.  When the Pod is finished, we'd like to recycle it and give it to a new Pod.  The existing Allocate (and the rest of the DPAPI) actually works great for us, we just need to be able to recycle the HW in an orderly way when the Pod it was Allocated to is gone.

Ed

[Vishnu Kannan]

Hey Ed,

Can you recycle as part of a subsequent Allocate() call, or are you looking to optimize pod startup latency by recycling as soon as a pod disappears?

To view this discussion on the web visit https://groups.google.com/d/msgid/kubernetes-wg-resource-management/CAFVSqg3QvL4HAOGMvHE3U1BbRyA5ShEsqwA98xrwRqJNhkMXHw%40mail.gmail.com.

[Ed Warnicke]

Ah… I had missed that :)  I’ll strongly suspect that will do it for me.  I’ll remove the agenda item.

Ed