Control plane init generates userdata larger than max user data size in AWS #510
Labels
kind/bug
Something isn't working
needs-priority
Indicates an issue or PR needs a priority assigning to it
needs-triage
Indicates an issue or PR lacks a `triage/foo` label and requires one.
What happened:
Creating a new cluster with rke2 provider v0.8.0, the AWSMachine fails to come up due to the error
It mostly seems to affect the first control plane node since the userdata for that node (control plane init) is larger.
Note that the limit is 25600 base64 encoded or around 16k unencoded. This is easily exceeded due to the certs included in the user data as well as rke2 config and any start up scripts needed.
What did you expect to happen:
First node comes up without error
How to reproduce it:
Create an RKE2 cluster using rke2 provider v0.8.0 and cluster-api AWS (CAPA)
Anything else you would like to add:
cloud-init supports gzipped userdata. If i manually replace the generated data with base64 encoded gzipped data, then the node comes up fine. I think the fix would be to always gzip cloud-init userdata.
Environment:
/etc/os-release
): ubuntu 22.04 LTSThe text was updated successfully, but these errors were encountered: