Gerrit
Tried in mini-lab successfully along with the metal-hammer: ``` metal_api_image_tag: pr-finalize-allocation-improvements metal_hammer_image_url: https://images.metal-stack.io/metal-hammer/pull-requests/57-wipe-disks/metal-hammer-initrd.img.lz4 ```
I am unable to reproduce this. Could be the #271 helps in this scenario, but don't know.
Here are related metal-api error logs: ```json {"level":"error","time":"2021-08-04T12:48:48.296Z","caller":"rest/middleware.go:88","message":"Rest Call","rqid":"5019de598aab8c4f96ea782160836959","remoteaddr":"10.20.4.3","method":"GET","uri":"/metal/v1/machine/abort-reinstall","route":"/metal/v1/machine/{id}","rqid":"5019de598aab8c4f96ea782160836959","remoteaddr":"10.20.4.3","method":"GET","uri":"/metal/v1/machine/abort-reinstall","route":"/metal/v1/machine/{id}","status":403,"content-length":55,"duration":0.000260831} {"level":"error","time":"2021-08-04T12:48:48.296Z","caller":"rest/middleware.go:100","message":"cannot get user from request","rqid":"5019de598aab8c4f96ea782160836959","remoteaddr":"10.20.4.3","method":"GET","uri":"/metal/v1/machine/abort-reinstall","route":"/metal/v1/machine/{id}","error":"Wrong HMAC found","got":"667c82a5692d7e1013a88b022af0f367e3d349860818ff8714b685bed3ae91e3","want":"4a2e464619017e291fd0113c5a884b89d825a5e8ea8fbd6b1a83eb5f1eab86c8"} {"level":"error","time":"2021-08-04T12:48:48.274Z","caller":"rest/middleware.go:88","message":"Rest Call","rqid":"bc0c64ff4f0ae73854a35da29d30d735","remoteaddr":"10.20.4.3","method":"POST","uri":"/metal/v1/machine/eff49e00-6ff4-11e9-8000-efbeaddeefbe/finalize-allocation","route":"/metal/v1/machine/{id}/finalize-allocation","useremail":"[email protected]","rqid":"bc0c64ff4f0ae73854a35da29d30d735","remoteaddr":"10.20.4.3","method":"POST","uri":"/metal/v1/machine/eff49e00-6ff4-11e9-8000-efbeaddeefbe/finalize-allocation","route":"/metal/v1/machine/{id}/finalize-allocation","status":422,"content-length":137,"duration":0.087437639} {"level":"error","time":"2021-08-04T12:48:48.274Z","caller":"service/service.go:96","message":"service error","rqid":"bc0c64ff4f0ae73854a35da29d30d735","remoteaddr":"10.20.4.3","method":"POST","uri":"/metal/v1/machine/eff49e00-6ff4-11e9-8000-efbeaddeefbe/finalize-allocation","route":"/metal/v1/machine/{id}/finalize-allocation","useremail":"[email protected]","operation":"finalizeAllocation","status":422,"error":"the machine \"eff49e00-6ff4-11e9-8000-efbeaddeefbe\" could not be enslaved into the vrf vrf731","service-caller":"machine-service.go:1635","resp":"the...
I tried to reconstruct what happened, here is what found: 1. Everything was according to plan, wiped primary disk and reinstalled OS 1. Then metal-hammer calls metal-core `/report`, returning an...
Yep, please make it conditional --> if headscale server is configured it fails, if it's not configured everything works as before
Good finding. It's definitely coming from the metal-api because the `vrfshared` property does not get set in the network response. Therefore it comes with `null` for this property which the...
Does this have any side effects I am not aware of? Solution looks so simple to me and our discussion in the past was fairly long. :thinking:
Why does it need to API breaking? Can't we just write to an alternative datastore for both endpoints if enabled?
You can use my fork, it's still working sine the days I've raised this pull request.
Sorry for slow response, your requested changes were already integrated into the chart, which is now located on https://github.com/metal-stack/helm-charts. It works at least with K8s v1.24 now. Thanks for your...