Failure to install go in e2e workflows

[lentzi90]

Which jobs are flaking?

I think this affects all e2e jobs since they all install go the same way.
Here is one example of a failing job: https://github.com/metal3-io/baremetal-operator/actions/runs/21616532469

Which tests are flaking?

Installation of go that happens before the tests actually start.

Since when has it been flaking?

First occurrence I have seen was on 1st February: https://github.com/metal3-io/baremetal-operator/actions/runs/21557005916/job/62115095091

Jenkins link

No response

Reason for failure (if possible)

There is some kind of issue downloading and extracting the archive. Probably it was corrupted during download.

+ /home/ubuntu/_work/baremetal-operator/baremetal-operator/hack/e2e/ensure_go.sh
+ MINIMUM_GO_VERSION=go1.24.12
+ verify_go_version
++ command -v go
+ '[' -x '' ']'
+ [[ linux-gnu == \l\i\n\u\x\-\g\n\u ]]
+ echo 'go not found, installing'
+ curl -sLo /tmp/go1.24.12.linux-amd64.tar.gz https://go.dev/dl/go1.24.12.linux-amd64.tar.gz
go not found, installing
+ sudo tar -C /usr/local -xzf /tmp/go1.24.12.linux-amd64.tar.gz

gzip: stdin: not in gzip format
tar: Child returned status 1
tar: Error is not recoverable: exiting now
Error: Process completed with exit code 2.

Anything else we need to know?

As a fix/workaround we could add a retry mechanism and file integrity check.

Label(s) to be applied

/kind flake
One or more /area label. See https://github.com/metal3-io/baremetal-operator/labels for the list of labels.

[Rozzii]

/triage accepted
/kind ci