aakashkt / ovn-kubernetes-charm Goto Github PK

Why doesn't gateway-physical-interface default to what you get from ip route | grep default?

As you requested https://github.com/juju-solutions/bundle-canonical-kubernetes/issues/364 here are some comments on the layers you have.

• I see you have two layers. You should have each layer on a separate git repo or in your case just create a single layer. The reason for this has to do with how charm build works. When you run a charm build the build process scans all the layers you are using (as stated in your layers.yaml) and performs a lookup on the http://interfaces.juju.solutions/ to get the source repository where each layer and/or interface is. Then the build process fetches locally the sources of all the layers and moves on with the build. For the fetch (git clone in our case) to work we should have a git repo for each layer. Also note that the layers have to be registered in http://interfaces.juju.solutions

• I see you have a run_command https://github.com/AakashKT/ovn-kubernetes-charm/blob/master/layers/openvswitch/reactive/openvswitch.py#L40 but you never check if the command failed.

• to get the public IP of the charm you should use https://pythonhosted.org/charmhelpers/api/charmhelpers.core.hookenv.html#charmhelpers.core.hookenv.unit_public_ip

• the basic layer you are using has a packaging option that will deploy the packages you request. So you do not need this: https://github.com/AakashKT/ovn-kubernetes-charm/blob/master/layers/openvswitch/reactive/openvswitch.py#L67 . Alternatively you can use the apt-layer like so: https://github.com/juju-solutions/layer-cwr/blob/master/layer.yaml

• You should avoid cloning openvswitch locally and installing it as you do here: https://github.com/AakashKT/ovn-kubernetes-charm/blob/master/layers/openvswitch/reactive/openvswitch.py#L80 You do not know when this repository will break, you do not have consistent deployments. Instead you should use apt packages apt-get install openvswitch-switch or at least package binaries into resources.

Was deploying the bunlde http://paste.ubuntu.com/25168507/ on AWS.

Some pods do not start. Here is a view of kubectl get po --all-namespaces

 NAMESPACE     NAME                                    READY     STATUS             RESTARTS   AGE
default       default-http-backend-44ppj              1/1       Running            0          28m
default       nginx-ingress-controller-2165r          0/1       CrashLoopBackOff   18         28m
kube-system   heapster-v1.2.0.1-3929440214-25gsb      4/4       Running            0          26m
kube-system   kube-dns-4101612645-bcq53               2/4       CrashLoopBackOff   15         30m
kube-system   kubernetes-dashboard-3543765157-lck0f   0/1       CrashLoopBackOff   13         30m
kube-system   monitoring-influxdb-grafana-v4-13h6g    2/2       Running            0          30m

Here is the kubernetes-worker syslog: http://paste.ubuntu.com/25168517/
And here is the master's syslog: http://paste.ubuntu.com/25168528/

Here is the output of kubectl logs po/kube-dns-4101612645-bcq53 -n kube-system kubedns http://paste.ubuntu.com/25168569/
And here is the description of kubedns http://paste.ubuntu.com/25168572/

I see some error on the ovn-k8s-watcher.log http://paste.ubuntu.com/25168607/

Hi,

You might want to use open_port[1] to ask Juju to open the ports your charm is using. This way you do not have to tamper with each cloud's firewall.

Here are some examples of how open_port is used [2, 3]

[1] https://pythonhosted.org/charmhelpers/api/charmhelpers.core.hookenv.html#charmhelpers.core.hookenv.open_port

[2] https://jujucharms.com/docs/2.2/howto-charm-with-docker

[3] https://github.com/juju-solutions/layer-benchmark-gui/blob/master/reactive/benchmark_gui.py#L80