You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Configuration:
Ansible Control Machine (Kali 2018.4 VM), using VMWare Workstation (natt'ed network)
Pull of ansible-rpi-k8s-project of 11/7/18.
Since both issues described here (static routing issue) and CPU spike kube-apiserver might be related keeping issues together in one report for now.
Problem 1: Static Routing issue - SSH Kali host
Cannot ssh from Ansible Control Machine (192.168.80.135) to rpi-k8s-1 (192.168.1.100, statically assigned). ssh to WIFI (DHCP) address (15.15.15.147) works fine
Expected results:
ssh from 192.168.80.135 to 192.168.1.100 to work (Similar as to 15.15.15.147)
Problem 2: kube-apiserver process consumes CPU and freezes rpi-k8s-1 within minutes
After successful execution of ansible-playbook -i inventory playbooks/deploy.yml, kube-apiserver
(by resolving issue 1 & 2 listed below) process runs out of memory and freezes rpi-k8s-1 within minutes
Workaround: manually kill kube-apiserver process each few minutes
Expected results:
Expect rpi-k8s-1 to continue to function (cap memory usage of of kube-apiserver process)
…..
TASK [Gathering Facts] ********************************************************************************************
Tuesday 06 November 2018 13:03:26 -0800 (0:00:01.136) 0:00:50.157 ******
fatal: [rpi-k8s-2]: UNREACHABLE! => {"changed": false, "msg": "SSH Error: data could not be sent to remote host \"192.168.100.128\". Make sure this host can be reached over ssh", "unreachable": true}
fatal: [rpi-k8s-3]: UNREACHABLE! => {"changed": false, "msg": "SSH Error: data could not be sent to remote host \"192.168.100.129\". Make sure this host can be reached over ssh", "unreachable": true}
Configuration:
Ansible Control Machine (Kali 2018.4 VM), using VMWare Workstation (natt'ed network)
Pull of ansible-rpi-k8s-project of 11/7/18.
Since both issues described here (static routing issue) and CPU spike kube-apiserver might be related keeping issues together in one report for now.
Problem 1: Static Routing issue - SSH Kali host
Cannot ssh from Ansible Control Machine (192.168.80.135) to rpi-k8s-1 (192.168.1.100, statically assigned). ssh to WIFI (DHCP) address (15.15.15.147) works fine
Expected results:
ssh from 192.168.80.135 to 192.168.1.100 to work (Similar as to 15.15.15.147)
Problem 2: kube-apiserver process consumes CPU and freezes rpi-k8s-1 within minutes
After successful execution of ansible-playbook -i inventory playbooks/deploy.yml, kube-apiserver
(by resolving issue 1 & 2 listed below) process runs out of memory and freezes rpi-k8s-1 within minutes
Workaround: manually kill kube-apiserver process each few minutes
Expected results:
Expect rpi-k8s-1 to continue to function (cap memory usage of of kube-apiserver process)
Resolved Issue 1:
ssh to 15.15.15.147 (WIFI rpi-k8s-1) fails.
SOLVED: Updated ansible.cfg: ssh_args = -o IPQos=throughput
Resolved Issue 2:
ansible-playbook -i inventory playbooks/deploy.yml fails to contact slave nodes
Snippet error in output
SOLVED: Updated inventory/group_vars/rpi_k8s_slaves/all.yml:
ansible_ssh_common_args: '-o ProxyCommand="ssh -o IPQos=throughput -W %h:%p -q {{ ansible_user }}@{{ jumphost_ip }}"'
After resolving issue 1 & 2 : ansible-playbook -i inventory playbooks/deploy.yml completes successfully
Continued/existing problems:
Problem 1: Cannot ssh to rpi-k8s-1 (192.168.1.100, statically assigned)
Ssh to rpi-k8s-1 (15.15.15.147, wifi, dhcp) works fine
Problem 2: Kube-apiserver consumes 100%cpu every few minutes and freezes rpi-k8s-1
The text was updated successfully, but these errors were encountered: