[nightly CI] capo kubeadm sylva upgrade misc job failing continuously at calico-ready

This issue is about a symptom ("calico-ready unit not becoming ready"), but has no formally identified root cause. In fact in #2941 (closed) (an issue which is a backport issue of the issue here) we understood that any failure during node rolling update can result in this symptom.

⚠️ So this issue can be kept as a reference, to help us link with any specific "root cause" issues/MRs, but should not be used to create backport issues. ⚠️

In #2941 (closed) (an issue which is a backport issue of the issue here), we had identified one root-cause for this symptom, which is tracked specifically in capo-contrail-bgpaas failing to remove OpenStac... (#2998).

Initial MR description

☁capo 🚀kadm 🎬sylva‑upgrade‑from‑1.4.x 🛠ha,misc 🐧ubuntu jobs are continuously failing at calico-ready

Scheduled pipelines report

jobs:

Edited by Thomas Morin