This message was deleted.
# harvester
a
This message was deleted.
b
Yeah, this is a gap that the new built-in managed DHCP feature is intended to solve in Harvester 1.3: https://github.com/harvester/harvester/blob/master/enhancements/20230830-managed-dhcp.md
b
Oh amazing! Is it in the pre-release already?
b
It's being developed as an experimental addon, so it's not currently tied to the ISOs we generate as release candidates. It lives here: https://github.com/harvester/experimental-addons/ Seems like we haven't merged docs describing how to install it yet though, but if you're curious, this PR looks close to being merged, so it should be approximately correct anyhow 😅 https://github.com/harvester/docs/pull/510/files?short_path=997bed4#diff-997bed408141e8a878258da29cd973015be0328823da5e3c774b5617ad064954
b
haha. Nice, this is amazing! Could I ask a followup question? I can sort of approximate this with a cloud-init script that sends a quick API call to our DHCP server on boot to make the static mapping. On spindown, I could have the node execute something as well to remove its mapping from DHCP server, but is there a better way here? Is there a hook elsewhere within harvester/rancher/k8s that would be most appropriate to make an API call to the DHCP server I have running? It’s asimple oneliner curl call.
b
I don't think we have any no/low-code "hooks" for expressing automation like this in the Harvester cluster.
b
What about in rancher? A post start hook and a pre stop hook for the nodes? 😁
b
I spend all of my time in Harvester so sadly I'm not as familiar with what's possible with Rancher 😅 Maybe the #general channel might have a better idea, if someone else doesn't come into this thread to weigh in separately.
b
Thank you!
w
I love you guys! This solves so many head-aches for me - and single points of failures.
Can’t wait until 1.3 is released.
b
!