[REQUEST] Lame Duck Mode

[jordan-rash]

As an extension to #692, I would like to see the "lame duck" tag implemented.

Similar to NATs lame duck mode, it would cause the host to drain itself and block further deployments (from wash and wadm) to the host. This would allow for easier updates of the host in place.

Ref: https://wasmcloud.slack.com/archives/C04RVMJJH17/p1695735058510099

[connorsmith256]

Some questions on this one:

• once in lame duck mode, should the host refuse to process any new control interface commands, or is this mode "just' part of what's returned by host pings/inventory requests?
• once the host has drained its contents, will it shut down, or does it remain in lame duck mode?
  • if the host stays running, should there be a way to switch out of lame duck mode, or is this a terminal state?

[jordan-rash]

IMO:

• once lame duck mode is enabled, no "new" commands should be processed. All existing workloads should be shutdown nicely (up to some TTL and then killed)
• I think once it is shutdown, it would remain on and continue to not respond to request. Reason for this is,
  1. maybe I was just troubleshooting something and will take it out of lame duck mode once I am done or
  2. Doing some sort of load testing
• I def believe lame duck mode should be toggleable

[protochron]

We've chatted about this before, and I'd also like to see this as a feature. Conceptually I think it as being closer to Kubernetes' drain/cordon behavior or Nomad's node draining/eligibility feature.

[jordan-rash]

We've chatted about this before, and I'd also like to see this as a feature. Conceptually I think it as being closer to Kubernetes' drain/cordon behavior or Nomad's node draining/eligibility feature.

But Dan def thinks that the feature should have duck in the name...right, RIGHT?!?!?

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. If this has been closed too eagerly, please feel free to tag a maintainer so we can keep working on the issue. Thank you for contributing to wasmCloud!

[thomastaylor312]

@jordan-rash did you mean to close this?