Pods are ContainerCreating state for long time about kubernetes_the_easy_way

[hoeghh]

Yes I saw it. I have made an issue in Kelsey's repo kubernetes the hard way.

[hoeghh]

I have created a pastbin with all logs on this error:

https://pastebin.com/GW2kwcLG

[vikasgubbi]

Any plans to change LXC to Docker as of now.

[hoeghh]

Yes i will change to docker from containerd
I just need to get some time for it 😊

[hoeghh]

Hi,

Could you try and do a ./destroy.sh, git pull and try again with ./install.sh
I think i have it working now. @vikasgubbi

[vikasgubbi]

Hi,

install.sh struck at " Installing loadbalancer" for more than 80 minutes. vagrant-provition.log output below

==> k8s-loadbalancer-1: Mounting shared folders...
k8s-loadbalancer-1: /vagrant => /root/kubernetes_the_easy_way
==> k8s-loadbalancer-1: Running provisioner: file...
==> k8s-loadbalancer-1: Running provisioner: file...
==> k8s-loadbalancer-1: Running provisioner: shell...
k8s-loadbalancer-1: Running: /tmp/vagrant-shell20180305-9802-ahgkwt.sh
k8s-loadbalancer-1: Installing loadbalancer...

logs folder contents

-rw-r--r--. 1 root root 2716 Mar 5 00:08 cert.log
-rw-r--r--. 1 root root 0 Mar 5 00:08 data-encryption.log
-rw-r--r--. 1 root root 477 Mar 5 00:08 kube-configs.log
-rw-r--r--. 1 root root 75861 Mar 5 00:32 vagrant-provition.log
-rw-r--r--. 1 root root 148 Mar 5 00:32 kubedeploy.log
-rw-r--r--. 1 root root 120 Mar 5 00:32 kubedns.log
-rw-r--r--. 1 root root 149 Mar 5 00:32 post_deploy.log
-rw-r--r--. 1 root root 193 Mar 5 00:32 weavenet.log

Regards,Vikas

[hoeghh]

I added some extra echo to the script that installs the loadbalacer. I pushed it around the time you ran it. It is strange that you don't get more output, if it's the newest commit.

Could you check that you have the newest ? It should say something like downloading binaries...

[hoeghh]

And please edit the config file and set the number of load balancers to 1 only.

Remember to destroy before trying again.

[vikasgubbi]

Hi,

I modified it before running install.sh. My config file

export ETCD_COUNT=1
export ETCD_CPU=1
export ETCD_MEM=512
export MASTER_COUNT=1
export MASTER_CPU=1
export MASTER_MEM=1024
export WORKER_COUNT=2
export WORKER_CPU=2
export WORKER_MEM=1048
export LOADBALANCER_COUNT=1
export LOADBALANCER_CPU=1
export LOADBALANCER_MEM=512

Regards,Vikas

[hoeghh]

Did you do a git pull before trying?

After printing the last line you get, it should print out another line. It seems strange that that would fail a simple echo command.

The file scripts/install_loadbalancer.sh is not empty right?

[vikasgubbi]

Hi,

No its not empty

ls -ltr
total 44
-rwxrwxrwx. 1 root root 1850 Mar 4 23:50 generate_kube_config.sh
-rwxrwxrwx. 1 root root 810 Mar 4 23:50 generate_hosts.sh
-rwxrwxrwx. 1 root root 314 Mar 4 23:50 generated_data_encryption.sh
-rwxrwxrwx. 1 root root 3526 Mar 4 23:50 generate_certs.sh
-rwxrwxrwx. 1 root root 4010 Mar 4 23:50 install-worker.sh
-rwxrwxrwx. 1 root root 5001 Mar 4 23:50 install-master.sh
-rwxrwxrwx. 1 root root 34 Mar 4 23:50 install-loadbalancer.sh
-rwxrwxrwx. 1 root root 1966 Mar 4 23:50 install-etcd.sh
-rwxrwxrwx. 1 root root 1915 Mar 4 23:50 install-corosync.sh
-rwxrwxrwx. 1 root root 1167 Mar 4 23:50 post_deploy.sh
[root@jenkins scripts]# more install-loadbalancer.sh
echo "Installing loadbalancer..."

Regards,Vikas

[hoeghh]

The file should have many lines of code.

Please do a git pull and check that the file has grown.

[vikasgubbi]

Hi,

Getting below mentioned message in vagrant-provition.log file

k8s-loadbalancer-1: Unable to connect to the server: dial tcp 192.168.50.20:6443: getsockopt: no route to host
k8s-loadbalancer-1: Getting token and ca.crt for Traefik from api-server...
k8s-loadbalancer-1: the server doesn't have a resource type "sa"
k8s-loadbalancer-1: Unable to connect to the server: dial tcp 192.168.50.20:6443: getsockopt: no route to host
k8s-loadbalancer-1: Unable to connect to the server: dial tcp 192.168.50.20:6443: getsockopt: no route to host
k8s-loadbalancer-1: Created symlink from /etc/systemd/system/multi-user.target.wants/traefik.service to /etc/systemd/system/traefik.service.
k8s-loadbalancer-1: Redirecting to /bin/systemctl start traefik.service
k8s-loadbalancer-1: Cleaning up...

Regards,Vikas

[hoeghh]

Could you check if there was any errors on the master machines in the log? Its seems that the floating ip (192.168.50.20) is not created.

Its also in the vagrant-provition.log file (search for master-1).

Did you do a ./destroy.sh before running ./install.sh ? If not, then try ./destroy.sh and then ./install.sh again.

[hoeghh]

I found the error. I had removed a line in the Vagrantfile that i shouldnt.
Im fixing it now, and adding HA to the loadbalancers.

I make a post here when i have pushed the changes.

[hoeghh]

Ok, so i've fixed the problem with Corosync and i have tested the cluster with two masters and two loadbalancers.

Please test again, and see if it works.
If it works, you should be able to get the traefik dashboard at http://192.168.50.4:8080

Fixed with bf50f0b

If it works, please comment and close the issue :)

[vikasgubbi]

Hi,

Load Balancer successfully installed,closing the ticket.

Regards,Vikas