improve UI advice when user needs to delete the cluster

[medyagh]

here we confuse the user with two things

❗ You cannot change the memory size for an exiting minikube cluster. Please first delete the cluster.
and

😿 Failed to start ssh bare metal machine. Running "minikube delete -p foo" may fix it: config: please provide an IP address

we should had sticked with first advice

drewpca(pts/0):~% minikube start -p foo --memory 6GB
😄  [foo] minikube v1.17.1 on Debian rodete
✨  Using the ssh driver based on existing profile
❗  You cannot change the memory size for an exiting minikube cluster. Please first delete the cluster.
👍  Starting control plane node foo in cluster foo
🤦  StartHost failed, but will try again: config: please provide an IP address
😿  Failed to start ssh bare metal machine. Running "minikube delete -p foo" may fix it: config: please provide an IP address
 
❌  Exiting due to GUEST_PROVISION: Failed to start host: config: please provide an IP address
 
😿  If the above advice does not help, please let us know: 
👉  https://github.com/kubernetes/minikube/issues/new/choose
 

[BLasan]

@medyagh You mean we do not need the log Failed to start ssh bare metal machine. Running "minikube delete -p foo" may fix it: config: please provide an IP address right? I would like to open a PR for this

[BLasan]

/assign

[BLasan]

@medyagh btw I forgot to add the output I received after running the command without the changes done in #10473

minikube start -p foo --memory 3GB
😄  [foo] minikube v1.17.1 on Ubuntu 20.04
✨  Using the docker driver based on existing profile
❗  You cannot change the memory size for an exiting minikube cluster. Please first delete the cluster.
👍  Starting control plane node foo in cluster foo
🔄  Restarting existing docker container for "foo" ...
🐳  Preparing Kubernetes v1.20.2 on Docker 20.10.2 ...
🔎  Verifying Kubernetes components...
    ▪ Using image gcr.io/k8s-minikube/storage-provisioner:v4
🌟  Enabled addons: storage-provisioner, default-storageclass
🏄  Done! kubectl is now configured to use "foo" cluster and "default" namespace by default

I was not getting errors below

🤦  StartHost failed, but will try again: config: please provide an IP address
😿  Failed to start ssh bare metal machine. Running "minikube delete -p foo" may fix it: config: please provide an IP address

[rahulii]

/assign
I would like to give it a shot. Just wanted to know where should i begin searching the code!
/cc @medyagh

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue or PR as fresh with /remove-lifecycle stale
• Mark this issue or PR as rotten with /lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[k8s-triage-robot]

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue or PR as fresh with /remove-lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle rotten

[spowelljr]

Hi @rahulii & @BLasan, it's been a while since you've both assigned yourselves to the issue so I unassigned you both. If you're still interested in working on this feel free to reassign yourselves and I can provide assistance with the implementation if needed, thanks!

[Zanderax]

/assign
This is my first issue so please go easy on me if I get things wrong 🙂

If I understand the issue, it's redundant and confusing to have both of these warnings.
❗ You cannot change the memory size for an exiting minikube cluster. Please first delete the cluster.
and
😿 Failed to start ssh bare metal machine. Running "minikube delete -p foo" may fix it: config: please provide an IP address

The first warning comes from this line where if the memory flag or the memory size has been changed then the warning is generated. Note that memory size is not the only resource that has a warning like this, there are warnings for CPUs, disk size, and extra disk numbers.

The second warning comes from this line where if the host fails to start twice it fails with a dynamically generated message.

These two warning are reached from this line appears to be trying to re-generate the cluster config just after deleting the cluster, giving warning one. Then it tries to provision a new cluster leading to warning two.

In this case, does the first error message still make sense? If we are re-generating the config after deleting then is it still an issue if we change the memory size?

We could change updateExistingConfigFromFlags() to not log existing cluster conflict warnings since there is no existing cluster, we are just using the old cluster's config to generate a new config. We could add a new flag argument to indicate that there is no existing cluster or we could make a new function that re-generates a config without the assumption that there is an existing cluster.

I will have a crack at the code this week and any advice would be much appreciated since I probably made a lot of mistakes and would like to learn.

[medyagh]

@Zanderax please feel free to take a stab at it and make a PR !

[Zanderax]

Sorry, my state opened up from a 3-month lockdown last week so I forgot about this a little bit as I was indulging in the novelty of leaving my house. I'll try to get around to it soon.

[Zanderax]

Sorry I'm going to unassign myself, I didn't expect to be this busy after lockdown finished and I won't be able to get to this. Sorry about that :)

[e-tienne]

/assign
grabbing it since it's now unassigned

[spowelljr]

@e-tienne Are you still working on this? I unassigned this task from you, but if you are actively working on it please reassign yourself, thanks!

[maknop]

/assign

[maknop]

Looks like this occurs when the driver is set to ssh. I'm going to give this issue a try. 😄

[BrinthaSivakami]

/assign

[akstron]

/assign

[akstron]

In contrast to the discussion here: #10460 (comment), I found that the first statement was not because of deletion and recreation of the cluster.
The call stack between the two statement is pretty large, so passing a flag seems difficult. What I suggest is removing the below line from the 2nd statement.
Running "minikube delete -p foo" may fix it