docker deployment, stuck at ContainerCreating

[Martin Veltins]

Hi there!

I am trying to get lucida working on my machine, ubuntu 16.04, 8GB RAM, 2,3GHz, 4 cores, 200GB space

I installed docker as stated: https://docs.docker.com/engine/installation/

cloned the git repo from https://github.com/claritylab/lucida

changed the hostPath within  mongo-controller.yaml and qa-controller.yaml

and ran

sudo ./cluster_up_linux.sh

As 

kubectl get pod

did not return any erros, i ran

 start_services.sh

Getting the state of the containers via

kubectl get pod

results in

root@system-N24-25JU:~# kubectl get pod
NAME                         READY     STATUS              RESTARTS   AGE
asrmaster-controller-dox2k   0/1       ContainerCreating   0          49m
asrworker-controller-z3svn   0/1       ContainerCreating   0          49m
asrworker-controller-zvrms   0/1       ContainerCreating   0          49m
ca-controller-5pdwi          0/1       ContainerCreating   0          49m
dig-controller-apulk         0/1       ContainerCreating   0          49m
face-controller-ub3lb        0/1       ContainerCreating   0          49m
imc-controller-rqfyw         0/1       ContainerCreating   0          49m
imm-controller-dupsl         0/1       ContainerCreating   0          49m
memcached-controller-l9b3f   1/1       Running             0          49m
mongo-controller-rtucu       0/1       ContainerCreating   0          49m
qa-controller-kdpt6          0/1       ContainerCreating   0          49m
web-controller-otaa2         0/1       ContainerCreating   0          49m

While setting everything up, i did this more than once.
In order to reset everything i stopped and deleted all docker container via

docker stop $(docker ps -a -q)
docker rm $(docker ps -a -q)

and rebooted the system.


In order to find the caus of these containers beeing stuck, i ran, e.g.:

kubectl describe pod asrmaster-controller-dox2k

 
[ ..... ]

 Status:        Pending

[ ..... ]

    State:            Waiting
      Reason:            ContainerCreating
    Ready:            False
    Restart Count:        0
    Environment Variables:    <none>

 

Well, this did not help me very much ..... .  Any ideas what went wrong?

 

[Martin Veltins]

Oh, btw!

I just tried to get some more information via (from /tools/deploy) :

docker ps | grep <controller_name> followed by docker exec -it <running_container_id> bash to go inside the running containers.

 

root@system:~# docker ps | grep imc-controller-rqfyw
7fad6bd48839        gcr.io/google_containers/pause-amd64:3.0               "/pause"                 36 minutes ago      Up 36 minutes                           k8s_POD.1d7605ff_imc-controller-rqfyw_default_2a6632ff-f786-11e6-aa4e-80fa5b337733_5fab4968
root@system:~# docker exec -it 7fad6bd48839 bash
rpc error: code = 2 desc = oci runtime error: exec failed: container_linux.go:247: starting container process caused "exec: \"bash\": executable file not found in $PATH"

Ok, as you see, this did not help me either .... .

[Martin Veltins]

Is there no edite mode here?!

It took way more time than i did expect ... .
I got all containers up and running

NAME                         READY     STATUS    RESTARTS   AGE

asrmaster-controller-dox2k   1/1       Running   0          1h
asrworker-controller-z3svn   1/1       Running   0          1h
asrworker-controller-zvrms   1/1       Running   0          1h
ca-controller-5pdwi          1/1       Running   0          1h
dig-controller-apulk         1/1       Running   0          1h
face-controller-ub3lb        1/1       Running   0          1h
imc-controller-rqfyw         1/1       Running   0          1h
imm-controller-dupsl         1/1       Running   0          1h
memcached-controller-l9b3f   1/1       Running   0          1h
mongo-controller-rtucu       1/1       Running   0          1h
qa-controller-kdpt6          1/1       Running   0          1h
web-controller-otaa2         1/1       Running   0          1h

Heading to http://localhost:30000
keeps spinning and spinning ....
After a while Firefox imforms me about the webpage not being available.
And the apache logs are empty .... .



root@system:~# docker ps | grep  web-controller-otaa2
e418a4561367        claritylab/lucida:v2.2                                 "/bin/sh -c 'cd $L..."   21 minutes ago      Up 21 minutes                           k8s_web.fbf77bfb_web-controller-otaa2_default_2ab7c32b-f786-11e6-aa4e-80fa5b337733_e2f4577b
2a0b879a58b2        gcr.io/google_containers/pause-amd64:3.0               "/pause"                 About an hour ago   Up About an hour                        k8s_POD.11e10785_web-controller-otaa2_default_2ab7c32b-f786-11e6-aa4e-80fa5b337733_8faee1b4
root@marlies-N24-25JU:~# docker exec -it e418a4561367 bash
root@web-controller-otaa2:/usr/local/lucida/lucida# cd /usr/local/lucida/lucida/commandcenter/apache/logs/
root@web-controller-otaa2:/usr/local/lucida/lucida/commandcenter/apache/logs# ls
root@web-controller-otaa2:/usr/local/lucida/lucida/commandcenter/apache/logs#




 





[Martin Veltins]

Restarting the computer and all services today lead to lucida being accessible via webbrowser.

Still got some problems left, but they are not related to this topic, trying to close it for now,
BR