`koff` ¶

koff is a command-line tool that allows you to process kubernetes resources in yaml or json format, from either file or piped input.
It reads input, performs the specific filter operations based on the flags and arguments (if provided), and writes the output in either tabular (as default), json or yaml format.

$ kubectl get pod,svc,ep -o yaml | koff
NAME                                      READY   STATUS      RESTARTS   AGE
pod/postgresql-1-2gxpm                    1/1     Running     0          15m
pod/postgresql-1-deploy                   0/1     Completed   0          16m
pod/rails-postgresql-example-1-build      0/1     Completed   0          16m
pod/rails-postgresql-example-1-deploy     0/1     Completed   0          11m
pod/rails-postgresql-example-1-hook-pre   0/1     Completed   0          11m
pod/rails-postgresql-example-1-v89ph      1/1     Running     0          11m

NAME                               TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)    AGE
service/postgresql                 ClusterIP   172.30.58.223   <none>        5432/TCP   16m
service/rails-postgresql-example   ClusterIP   172.30.114.85   <none>        8080/TCP   16m

NAME                                 ENDPOINTS           AGE
endpoints/postgresql                 10.129.3.75:5432    16m
endpoints/rails-postgresql-example   10.128.2.184:8080   16m

How it works?¶

Via piped input:

$ cat resources.yaml | koff get pod/postgresql-1-2gxpm svc/postgresql
NAME                     READY   STATUS    RESTARTS   AGE
pod/postgresql-1-2gxpm   1/1     Running   0          15m

NAME                 TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)    AGE
service/postgresql   ClusterIP   172.30.58.223   <none>        5432/TCP   16m

Referencing the file to use via koff use <resources>.yaml before executing koff:

$ koff use resources.yaml
$ koff get pod/postgresql-1-2gxpm svc/postgresql
NAME                     READY   STATUS    RESTARTS   AGE
pod/postgresql-1-2gxpm   1/1     Running   0          15m

NAME                 TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)    AGE
service/postgresql   ClusterIP   172.30.58.223   <none>        5432/TCP   16m

Using an etcd snapshot to get the Kubernetes object contained in it via koff etcd inspect:

$ koff etcd inspect etcd_snap.db /kubernetes.io/pods/openshift-etcd/testocp-bqgqk-master-0-debug -o yaml
apiVersion: v1
kind: Pod
metadata:
  annotations:
    debug.openshift.io/source-container: container-00
    debug.openshift.io/source-resource: /v1, Resource=nodes/testocp-bqgqk-master-0
  creationTimestamp: "1995-01-17T00:0:00Z"
  ...
  ...

$ koff etcd inspect etcd_snap.db /kubernetes.io/pods/openshift-etcd/testocp-bqgqk-master-0-debug | koff
NAME                                    READY   STATUS    RESTARTS   AGE
pod/testocp-bqgqk-master-0-debug        0/1     Pending   0          22h

Dealing with Custom Resources

To return a custom resource table in its default tabular format it's needed to add the respective CustomResourceDefinition yaml manifest into the path ~/.koff/customresourcedefinition.
As default behaviour, if the CustomResourceDefinition for the custom resource is not found the resulting table will have three columns only: NAMESPACE, NAME, CREATED AT.

$ kubectl get clusteroperator etcd -o yaml | koff 
NAME                                       CREATED AT
clusteroperator.config.openshift.io/etcd   2023-05-15T01:53:20

$ kubectl get crd clusteroperators.config.openshift.io -o yaml > ~/.koff/customresourcedefinitions/clusteroperators.config.openshift.io.yaml

$ kubectl get clusteroperator etcd -o yaml | koff
NAME                                       VERSION   AVAILABLE   PROGRESSING   DEGRADED   SINCE
clusteroperator.config.openshift.io/etcd   4.9.59    True        False         False      2d

koff ¶

How it works?¶

`koff` ¶