Failed to connect to : Connection refused

[Humberto Cunha Crispim]

After creating the VMS in the OCI, this loop appears and does not go to the next step

null_resource.wait_for_kube_apiserver (local-exec): curl: (7) Failed to connect to 132.226.252.238 port 6443 after 78 ms: Connection refused

[antonsatskyi]

same here

[RMorgado]

Same here :(

[marlon]

same here :(

[Andrei]

same here :(

[Weskhel]

same here :(

[JandersonFB]

same here :'(

[Jérôme Petazzoni]

Hi! Please try again with the latest version (make sure to git pull!) as I just pushed a bunch of updates that should improve the process a lot.

Also, now Terraform will show the cloud-init progress; I hope it will give a better idea about what's going on, especially in case of problems!

[JandersonFB]

All nodes started:

but, weave nets Crash:

[RMorgado]

I'm using the latest version and I have this error:

[Jérôme Petazzoni]

@JandersonFB Sorry about that - I got the wrong Weave YAML URL. I was using https://github.com/weaveworks/weave/releases/download/v2.8.1/weave-daemonset-k8s.yaml instead of https://github.com/weaveworks/weave/releases/download/v2.8.1/weave-daemonset-k8s-1.11.yaml. I've fixed the Terraform configuration and hopefully it should work now. Make sure you have the latest commits and after recreating the VMs it should work!

[Jérôme Petazzoni]

@RMorgado ah that's an interesting error. Do you get the same error for all 4 instances or just the first one? I wonder if that could be a transient error; i.e. if you try again after a while it will work? (Perhaps delete the VM and recreate it with Terraform; or use terraform taint? Let me know if you need help for that!)

[RMorgado]

Yes @jpetazzo , I have the error in all instances.
I've tested it several times and on different days.
I entered an instance and ran the command "kubectl get nodes " and I got this error:

I can send more information if you need it.
Thank you for your time and patience

[Hebert Viana]

Hello, first thanks for the work @jpetazzo, I'm having the same problem as @RMorgado.

I'm trying to identify the reason, at night I'll perform new tests if I get a solution I'll update here.

[Hebert Viana]

@jpetazzo and @RMorgado,

To solve the problem, I upgraded the provider version in provider.tf or oracle/oci to the latest version "4.102.0" (only that didn't work for me, I was in version 0.14.8 of terraform), I did the upgrade from terraform to v1.3.6 and successfully run terraform.

Apply complete! Resources: 5 added, 1 changed, 4 destroyed.

[RMorgado]

Thank you for your help @hebertviana !!!
I made the changes you said and updated terraform to the latest version and it worked!!

But I still get this error, I'll have to investigate to see if it's firewall issues

If I enter one of the nodes everything is ok!

[Hebert Viana]

@RMorgado,

Uhuu, glad it worked.

see if I managed to run this script to release the ports on the firewall, from what I saw here it is in /etc/cloud init/scripts

1-allow-inbound-traffic.sh

[PauloBigooD]

@hebertviana
Could you explain better how this configuration of the /etc/cloud init/scripts script was performed

1-allow-inbound-traffic.sh

I did a search here locally but didn't find anything related

Here to solve the problem I commented the following lines of the main.tf file

provisioner "remote-exec" {
inline = [
"tail -f /var/log/cloud-init-output.log &",
"cloud-init status --wait >/dev/null",
]
}

But I know this is just a palliative solution

[Jérôme Petazzoni]

Hi everyone! It looks like an extra firewalling rule had been added to the Oracle images. I removed that rule (in commit 0a82500) and it looks like it solved it. Let me know if it works for you!

[Alesson Viana]

Hey I updated my fork, and I compared the files, and seems like the same. but I'm still getting the same error