Skip to content
This repository has been archived by the owner on Feb 5, 2020. It is now read-only.

Installing tectonic for AWS - Not completing #12

Closed
tnkumar opened this issue Dec 22, 2016 · 10 comments
Closed

Installing tectonic for AWS - Not completing #12

tnkumar opened this issue Dec 22, 2016 · 10 comments

Comments

@tnkumar
Copy link

tnkumar commented Dec 22, 2016

Issue Report Template

Tectonic Version

tectonic 1.4.5

Environment

Windows 7 Enterprise
Intel i5, 8 GB RAM, 64-bit

What hardware/cloud provider/hypervisor is being used with Tectonic?
AWS

Expected Behavior

Actual Behavior

image
image

Reproduction Steps

  1. ...
  2. ...

Other Information

Feature Request

Environment

What hardware/cloud provider/hypervisor is being used with Tectonic?

Desired Feature

Other Information

@mfburnett
Copy link
Contributor

Hi @tnkumar, thanks for filing this issue! To help you troubleshoot, have you checked to make sure that you have the correct privileges for your AWS credentials?

@tnkumar
Copy link
Author

tnkumar commented Dec 28, 2016

Yes @mfburnett . I am the root administrator for the AWS account. Any other suggestions?

@sym3tri
Copy link
Contributor

sym3tri commented Dec 28, 2016

Hi @tnkumar can you provide us more information to help troubleshoot?

  1. I see 401 errors in your console log. Any idea why those would be appearing with your credentials? Perhaps they expired?
  2. What region and instance types did you choose?
  3. Can you share the output from AWS CloudFormation (the link is at the top of the installer screenshot you shared)?
  4. If you are able to SSH to the nodes in the cluster, what do you see for the kubelet statuses?
    systemctl status kubelet
    journalctl -u kubelet

@webner
Copy link

webner commented Jan 11, 2017

I had the same problem with the aws region eu-west-2. It seems that the regions are hard coded in kubernetes and london will be added with version 1.5.

@naphthalene
Copy link

I ran into this problem, see #19

You should be able to just scp the assets zip, explode it and run bootkube start. It appears to be waiting for the apiserver to start, but that's done by bootkube.

@dghubble
Copy link
Member

It is not actually waiting for the Kubernetes control plane, it waits for the Kubelet to be running, as resolved behind the ELB. The messaging was wrong. You can connect to your EC2 instances and verify the kubelet service is running. If that's ok, check the ELB or local network restrictions. Skipping over the validation can mask other issues.

@Nourid5
Copy link

Nourid5 commented Feb 22, 2017

Hi ... I am trying to install tectonic on bare metal, however I don't know how and where to run the matchboxes and get the informations for the "matchbox address" panel. Thanks

@dghubble
Copy link
Member

@Nourid5 this issue is about AWS. Please open a separate issue.

@mfburnett
Copy link
Contributor

Hi @tnkumar, @naphthalene, and @webner - we recently released Tectonic v1.5.4-tectonic.1, which makes a bunch of improvements over Tectonic 1.4.5, including removing the scp step and adding HA by default for AWS (and bare metal). You can give it a try here to see if this resolves some of the issues you were having: https://coreos.com/tectonic/releases/

@mfburnett
Copy link
Contributor

Going to close this issue due to inactivity. We recently released Tectonic v1.6.2-tectonic.1 with a ton of improvements - would recommend you all check it out! Feel free to open back up this issue with any comments or additional questions.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Development

No branches or pull requests

7 participants