Skip to content

Instantly share code, notes, and snippets.

Embed
What would you like to do?
My Openshift Cheatsheet

My Openshift Cheatsheet

Examine the cluster quota defined for the environment:

$ oc describe AppliedClusterResourceQuota

Install pkgs using yum in a Dockerfile

# Install Runtime Environment
RUN set -x && \ 2
    yum clean all && \
    REPOLIST=rhel-7-server-rpms,rhel-7-server-optional-rpms,rhel-7-server-thirdparty-oracle-java-rpms \
    INSTALL_PKGS="tar java-1.8.0-oracle-devel" && \
    yum -y update-minimal --disablerepo "*" --enablerepo ${REPOLIST} --setopt=tsflags=nodocs \
      --security --sec-severity=Important --sec-severity=Critical && \
    yum -y install --disablerepo "*" --enablerepo ${REPOLIST} --setopt=tsflags=nodocs ${INSTALL_PKGS} && \
    yum clean all

Creates a service to point to an external service addr (DNS or IP)

oc create service externalname myservice \
    --external-name myhost.example.com

A typical service creates endpoint resources dynamically, based on the selector attribute of the service. The oc status and oc get all commands do not display these resources. You can use the oc get endpoints command to display them.

If you use the oc create service externalname --external-name command to create a service, the command also creates an endpoint resource that points to the host name or IP address given as argument.

If you do not use the --external-name option, it does not create an endpoint resource. In this case, you need to use the oc create -f command and a resource definition file to explicitly create the endpoint resources.

If you create an endpoint from a file, you can define multiple IP addresses for the same external service, and rely on the OpenShift service load-balancing features. In this scenario, OpenShift does not add or remove addresses to account for the availability of each instance. An external application needs to update the list of IP addresses in the endpoint resource.

Patching a DeploymentConfig from the CLI

  • this example removes an config attribute using JSON path
oc patch dc/mysql --type=json \
    -p='[{"op":"remove", "path": "/spec/strategy/rollingParams"}]'
  • this example cnhage an existing attribute value using JSON format
oc patch dc/mysql --patch \
    '{"spec":{"strategy":{"type":"Recreate"}}}'

Creating a Custom template by exporting existing resources

The oc export command can create a resource definition file by using the --as-template option. Without the --as-template option, the oc export command only generates a list of resources. With the --as-template option, the oc export command wraps the list inside a template resource definition. After you export a set of resources to a template file, you can add annotations and parameters as desired.

The order in which you list the resources in the oc export command is important. You need to export dependent resources first, and then the resources that depend on them. For example, you need to export image streams before the build configurations and deployment configurations that reference those image streams.

oc export is,bc,dc,svc,route --as-template > mytemplate.yml

Depending on your needs, add more resource types to the previous command. For example, add secret before bc and dc. It is safe to add pvc to the end of the list of resource types because a deployment waits for persistent volume claim to bind.

The oc export command does not generate resource definitions that are ready to use in a template. These resource definitions contain runtime information that is not needed in a template, and some of it could prevent the template from working at all. Examples of runtime information are attributes such as status, creationTimeStamp, image, and tags, besides most annotations that start with the openshift.io/generated-by prefix.

Some resource types, such as secrets, require special handling. It is not possible to initialize key values inside the data attribute using template parameters. The data attribute from a secret resource needs to be replaced by the stringData attribute and all key values need to be unencoded.

Logging Aggregation throubleshooting

Process a template, create a new binary build to customize something and them change the DeploymentConfig to use the new Image...

oc process openshift//datagrid72-basic | oc create -f -

oc new-build --name=customdg -i openshift/jboss-datagrid72-openshift:1.0 --binary=true --to='customdg:1.0'
oc set triggers dc/datagrid-app --from-image=openshift/jboss-datagrid72-openshift:1.0 --remove
oc set triggers dc/datagrid-app --from-image=customdg:1.0 -c datagrid-app

List only paramaters of a given template file definition

oc process -f mytemplate.yaml --parameters

Copy file content from a specific image to local file system

docker run registry.access.redhat.com/jboss-datagrid-7/datagrid72-openshift:1.0 /bin/sh -c 'cat /opt/datagrid/standalone/configuration/clustered-openshift.xml' > clustered-openshift.xml

set the default storage-class

oc patch storageclass glusterfs-storage -p '{"metadata": {"annotations":{"storageclass.kubernetes.io/is-default-class":"true"}}}'

Change Default response timeout for a specific route:

oc annotate route <route_name> --overwrite haproxy.router.openshift.io/timeout=10s

Add a nodeSelector on RC ou DC

oc patch dc|rc <dc_name> -p "spec:                                                                                         
  template:     
    spec:
      nodeSelector:
        region: infra"

Binary Builds

oc new-build --binary=true --name=ola2 --image-stream=redhat-openjdk18-openshift --to='mycustom-jdk8:1.0'
oc start-build ola2 --from-file=./target/ola.jar --follow
oc new-app 

Turn off/on DC triggers to do a batch of changes without spam many deployments

oc rollout pause dc <dc name>
oc rollout resume dc <dc name> 

get a route URL using OC

http://$(oc get route nexus3 --template='{{ .spec.host }}')

Using Nexus repo manager to store deployment artifacts

Maven uses settings.xml in $HOME/.m2 for configuration outside of pom.xml:

<?xml version="1.0"?>
<settings>
  <mirrors>
    <mirror>
      <id>Nexus</id>
      <name>Nexus Public Mirror</name>
      <url>http://nexus-opentlc-shared.cloudapps.na.openshift.opentlc.com/content/groups/public/</url>
      <mirrorOf>*</mirrorOf>
    </mirror>
  </mirrors>
  <servers>
  <server>
    <id>nexus</id>
    <username>admin</username>
    <password>admin123</password>
  </server>
</servers>
</settings>

Maven can automatically store artifacts using -DaltDeploymentRepository parameter for deploy task:

mvn deploy -DskipTests=true \
-DaltDeploymentRepository= nexus::default::http://nexus3.nexus.svc.cluster.local:8081/repository/releases

to update a DeploymentConfig in order to change the Docker Image used by a specific container

oc project <project>
oc get is

# creates an ImageStream from a Remote Docker Registry image
oc import-image <image name> --from=docker.io/<imagerepo>/<imagename> --all --confirm

oc get istag

OC_EDITOR="vim" oc edit dc/<your_dc>

    spec:
      containers:
      - image: docker.io/openshiftdemos/gogs@sha256:<the new image digest from Image Stream>
        imagePullPolicy: Always

BuildConfig with Source pull secrets

oc secrets new-basicauth gogs-basicauth --username=<your gogs login> --password=<gogs pwd>
oc set build-secret --source bc/tasks gogs-basicauth

Adding a volume in a given DeploymentConfig

oc set volume dc/myAppDC --add --overwrite --name....

Create a configmap file and mount as a volume on DC

oc create configmap myconfigfile --from-file=./configfile.txt
oc set volumes dc/printenv --add --overwrite=true --name=config-volume --mount-path=/data -t configmap --configmap-name=myconfigfile

create a secret via CLI

oc create secret generic mysec --from-literal=app_user=superuser --from-literal=app_password=topsecret
oc env dc/printenv --from=secret/mysec
oc set volume dc/printenv --add --name=db-config-volume --mount-path=/dbconfig --secret-name=printenv-db-secret

Configure Liveness/Readiness probes on DCs

oc set probe dc cotd1 --liveness -- echo ok
oc set probe dc/cotd1 --readiness --get-url=http://:8080/index.php --initial-delay-seconds=2 

Create a new JOB

oc run pi --image=perl --replicas=1  --restart=OnFailure \
    --command -- perl -Mbignum=bpi -wle 'print bpi(2000)'

CRON JOB

oc run pi --image=perl --schedule='*/1 * * * *' \
    --restart=OnFailure --labels parent="cronjobpi" \
    --command -- perl -Mbignum=bpi -wle 'print bpi(2000)'

A/B Deployments - Split route trafic between services

oc expose service cotd1 --name='abcotd' -l name='cotd'
oc set route-backends abcotd --adjust cotd2=+20%
oc set route-backends abcotd cotd1=50 cotd2=50

to pull an image directly from red hat offcial docker registry

docker pull registry.access.redhat.com/jboss-eap-6/eap64-openshift

to validate a openshift/kubernates resource definition (json/yaml file) in order to find malformed/sintax problems

oc create --dry-run --validate -f openshift/template/tomcat6-docker-buildconfig.yaml

to get current user Barear Auth Token

oc whoami -t

to test Master API

curl -k -H "Authorization: Bearer <api_token>" https://<master_host>:8443/api/v1/namespaces/<projcet_name>/pods/https:<pod_name>:8778/proxy/jolokia/

# get pod memory via jmx
curl -k -H "Authorization: Bearer <api_token>" https://<master_host>:8443/api/v1/namespaces/<projcet_name>/pods/https:<pod_name>:8778/proxy/jolokia//read/java.lang:type\=Memory/HeapMemoryUsage | jq .

to login via CLI oc

oc login --username=tuelho --insecure-skip-tls-verify --server=https://master00-${guid}.oslab.opentlc.com:8443

### to login as Cluster Admin through master host
oc login -u system:admin -n openshift

to view the cluster roles and their associated rule sets in the cluster policy

oc describe clusterPolicy default

add a role to user

#local binding
oadm policy add-role-to-user <role> <username>

#cluster biding
oadm policy add-cluster-role-to-user <role> <username>

allow containers run with root user inside openshift

oadm policy add-scc-to-user anyuid -z default

for more details consult: https://docs.openshift.com/enterprise/3.1/admin_guide/manage_authorization_policy.html

to test a POD service locally

ip=`oc describe pod hello-openshift|grep IP:|awk '{print $2}'`
curl http://${ip}:8080

to access a POD container shell

oc exec -ti  `oc get pods |  awk '/registry/ { print $1; }'` /bin/bash

#new way to do the same:
oc rsh <container-name>

to edit an object/resource

oc edit <object_type>/<object_name>

#eg

oc edit dc/myDeploymentConfig

Ataching a new PersistentVolumeClaim to a DeploymentConfig

oc volume dc/docker-registry \
   --add --overwrite \
   -t persistentVolumeClaim \
   --claim-name=registry-claim \
   --name=registry-storage

Docker builder app creation

oc new-app --docker-image=openshift/hello-openshift:v1.0.6 -l "todelete=yes"

To create an app using a template (eap64-basic-s2i): Ticketmonster demo

oc new-app javaee6-demo
oc new-app --template=eap64-basic-s2i -p=APPLICATION_NAME=ticketmonster,SOURCE_REPOSITORY_URL=https://github.com/jboss-developer/ticket-monster,SOURCE_REPOSITORY_REF=2.7.0.Final,CONTEXT_DIR=demo

STI app creation

oc new-app https://github.com/openshift/sinatra-example -l "todelete=yes"
oc new-app openshift/php~https://github.com/openshift/sti-php -l "todelete=yes"

To watch a build process log

oc get builds
oc logs -f builds/sti-php-1

To create application using Git repository at current directory:

$ oc new-app

To create application using remote Git repository and context subdirectory:

$ oc new-app https://github.com/openshift/sti-ruby.git \
    --context-dir=2.0/test/puma-test-app

To create application using remote Git repository with specific branch reference:

$ oc new-app https://github.com/openshift/ruby-hello-world.git#beta4

New App From Source Code

Build Strategy Detection

If new-app finds a Dockerfile in the repository, it uses docker build strategy Otherwise, new-app uses source strategy

To specify strategy, set --strategy flag to source or docker Example: To force new-app to use docker strategy for local source repository:

$ oc new-app /home/user/code/myapp --strategy=docker

to create a definition generated by oc new-app command based on S2I support

$ oc new-app https://github.com/openshift/simple-openshift-sinatra-sti.git -o json | \
   tee ~/simple-sinatra.json

To create application from MySQL image in Docker Hub:

$ oc new-app mysql

To create application from local registry:

$ oc new-app myregistry:5000/example/myimage

If the registry that the image comes from is not secured with SSL, cluster administrators must ensure that the Docker daemon on the OpenShift Enterprise nodes is run with the --insecure-registry flag pointing to that registry. You must also use the --insecure-registry=true flag to tell new-app that the image comes from an insecure registry.

To create application from stored template:

$ oc create -f examples/sample-app/application-template-stibuild.json
$ oc new-app ruby-helloworld-sample

To set environment variables when creating application for database image:

$ oc new-app openshift/postgresql-92-centos7 \
    -e POSTGRESQL_USER=user \
    -e POSTGRESQL_DATABASE=db \
    -e POSTGRESQL_PASSWORD=password

To output new-app artifacts to file, edit them, then create them using oc create:

$ oc new-app https://github.com/openshift/ruby-hello-world -o json > myapp.json
$ vi myapp.json
$ oc create -f myapp.json
  • To deploy two images in single pod:
$ oc new-app nginx+mysql

To deploy together image built from source and external image:

$ oc new-app \
    ruby~https://github.com/openshift/ruby-hello-world \
    mysql \
    --group=ruby+mysql

to export all the project's objects/resources as a single template:

$ oc export all --as-template=<template_name>

You can also substitute a particular resource type or multiple resources instead of all. Run $ oc export -h for more examples

  • to create a new project using oadm and defining an admin user
$ oadm new-project instant-app --display-name="instant app example project" \
    --description='A demonstration of an instant-app/template' \
    --node-selector='region=primary' --admin=andrew

to create an app using oc CLI based on a template

$ oc new-app --template=mysql-ephemeral --param=MYSQL_USER=mysqluser,MYSQL_PASSWORD=redhat,MYSQL_DATABASE=mydb,DATABASE_SERVICE_NAME=database

to see a list of env vars defined in a DeploymentConfig object

$ oc env dc database --list
# deploymentconfigs database, container mysql
MYSQL_USER=***
MYSQL_PASSWORD=***
MYSQL_DATABASE=***

to manage enviorenmet variables in different ose objects types.

The first adds, with value /data. The second updates, with value /opt.

$ oc env dc/registry STORAGE=/data
$ oc env dc/registry --overwrite STORAGE=/opt

To unset environment variables in the pod templates:

$ oc env <object-selection> KEY_1- ... KEY_N- [<common-options>]

The trailing hyphen (-, U+2D) is required.

This example removes environment variables ENV1 and ENV2 from deployment config d1:

$ oc env dc/d1 ENV1- ENV2-

This removes environment variable ENV from all replication controllers:

$ oc env rc --all ENV-

This removes environment variable ENV from container c1 for replication controller r1:

To list environment variables in pods or pod templates:

$ oc env rc r1 --containers='c1' ENV-

This example lists all environment variables for pod p1:

$ oc env <object-selection> --list [<common-options>]
$ oc env pod/p1 --list

to apply some change (patch)

oc patch dc/<dc_name> \
   -p '{"spec":{"template":{"spec":{"nodeSelector":{"nodeLabel":"logging-es-node-1"}}}}}'

to apply a vlome storage

oc volume dc/<dc_name> \
          --add --overwrite --name=<volume_name> \
          --type=persistentVolumeClaim --claim-name=<claim_name>

to make a node unschedulable in a cluster

oadm manage node <nome do node > --schedulable=false

to create a registry with storage-volume mounted on host

oadm registry --service-account=registry \
    --config=/etc/origin/master/admin.kubeconfig \
    --credentials=/etc/origin/master/openshift-registry.kubeconfig \
    --images='registry.access.redhat.com/openshift3/ose-${component}:${version}' \
    --mount-host=<path> --selector=meuselector

to export all resources from a project/namespace as a template

oc export all --as-template=<template_name>

to create a build from a Dockerfile

# create the build
cat ./path/to/your/Dockerfile | oc new-build --name=build-from-docker --binary --strategy=docker -l app=app-from-custom-docker-build  -D -

#if you need to give some input to your Docker Build process
oc start-build build-from-docker --from-dir=. --follow

# create an OSE app from the docker build image
oc new-app app-from-custom-docker-build -l app=app-from-custom-docker-build

oc expose service app-from-custom-docker-build

to copy files to/from a POD

#Ref: https://docs.openshift.org/latest/dev_guide/copy_files_to_container.html

oc rsync /home/user/source devpod1234:/src

oc rsync devpod1234:/src /home/user/source

Cluster nodes CleanUp

  1. Desliga todos os containers que vc não tá usando no seu ambiente do openshift
  2. Executa em todos os nodes e master o comando: docker rm $(docker ps -a -q)
  3. Remove todas as imagens de todos os nodes e master. Para isso loga em cada uma delas via ssh e remove as imagens usando docker rmi . Pega as imagens que começa com o ip do registry 172.30...
  4. configurar GC: https://docs.openshift.com/enterprise/3.1/admin_guide/garbage_collection.html

##Tips

  • internal DNS name of ose/kubernetes services
  • follows the pattern <service-name>.<project>.svc.cluster.local
Object Type Example
Default <pod_namespace>.cluster.local
Services .<pod_namespace>.svc.cluster.local
Endpoints ..endpoints.cluster.local

"he only caveat to this, is that if we are using the multi-tenant OVS networking plugin, our cluster administrators will have to make visible our ci project to all other projects:" Ref: https://blog.openshift.com/improving-build-time-java-builds-openshift/

$ oadm pod-network make-projects-global ci

Adjust Master Log Level

To adjust openshift-master log level, edit following line of /etc/sysconfig/atomic-openshift-master from master VM:

OPTIONS=--loglevel=4

To make changes valid, restart atomic-openshift-master service:

$ sudo -i systemctl restart atomic-openshift-master.service

In node machine, to provide filtered information:

# journalctl -f -u atomic-openshift-node

Enable EAP clustering/replication

Make sure that your default service account has sufficient privileges to communicate with the Kubernetes REST API. Add the view role to serviceaccount for the project:

$ oc policy add-role-to-user view system:serviceaccount:$(oc project -q):default

Examine the first entry in the log file:

Service account has sufficient permissions to view pods in kubernetes (HTTP 200). Clustering will be available.

OCP Internal VIP failover for Routers running on Infra nodes

oc adm ipfailover ipf-ha-router 
    --replicas=2 --watch-port=80 \
    --selector="region=infra" \
    --virtual-ips="x.0.0.x" \
    --iptables-chain="INPUT" \
    --service-account=ipfailover --create

Creating a new Template

  • Common strategy for building template definitions:
  • Use oc new-app and oc expose to manually create resources application needs
  • Test to make sure resources work as expected
  • Use oc export with -o json option to export existing resource definitions
  • Merge resource definitions into template definition file
  • Add parameters
  • Test resource definition in another project

JSON syntax errors are not easy to identify, and OpenShift is sensitive to them, refusing JSON files that most browsers would accept as valid. Use jsonlint -s from the python-demjson package, available from EPEL, to identify syntax issues in a JSON resource definition file.

  • Use oc new-app with -o json option to bootstrap your new template definition file
oc new-app -o json openshift/hello-openshift > hello.json 
  • Converting the Resource Definition to a Template
  • Change kind attribute from List to Template
  • Make two changes to metadata object:
  • Add name attribute and value so template has name users can refer to
  • Add annotations containing description attribute for template, so users know what template is supposed to do.
  • Rename items array attribute as objects

Working with Templates

  • to list all parameters from mysql-persistent template:
$ oc process --parameters=true -n openshift mysql-persistent
  • Customizing resources from a preexisting Template

Example:

$ oc export -o json
	-n openshift mysql-ephemeral > mysql-ephemeral.json
... change the mysql-ephemeral.json file ...
$ oc process -f mysql-ephemeral.json \
	-v MYSQL_DATABASE=testdb,MYSQL_USE=testuser,MYSQL_PASSWORD=
	> testdb.json
$ oc create -f testdb.json

oc process uses the -v option to provide parameter values, while oc new-app command uses the -p option.

Create Definition Files for Volumes

ssh master00-$guid
mkdir /root/pvs
export volsize="5Gi"
for volume in pv{1..25}; \
do \
cat << EOF > /root/pvs/${volume}.yaml
apiVersion: v1
kind: PersistentVolume
metadata:
  name: ${volume} 
spec:
  capacity:
    storage: ${volsize} 
  accessModes:
  - ReadWriteOnce 
  nfs: 
    path: /var/export/pvs/${volume} 
    server: 192.168.0.254 
  persistentVolumeReclaimPolicy: Recycle 
EOF
     echo "Created def file for ${volume}"; \
done
export volsize="10Gi"
for volume in pv{26..50}; \
do \
cat << EOF > /root/pvs/${volume}.yaml
apiVersion: v1
kind: PersistentVolume
metadata:
  name: ${volume} 
spec:
  capacity:
    storage: ${volsize} 
  accessModes:
  - ReadWriteOnce 
  nfs: 
    path: /var/export/pvs/${volume} 
    server: 192.168.0.254 
  persistentVolumeReclaimPolicy: Recycle 
EOF
     echo "Created def file for ${volume}"; \
done
export volsize="1Gi"
for volume in pv{51..100}; \
do \
cat << EOF > /root/pvs/${volume}.yaml
apiVersion: v1
kind: PersistentVolume
metadata:
  name: ${volume} 
spec:
  capacity:
    storage: ${volsize} 
  accessModes:
  - ReadWriteOnce 
  nfs: 
    path: /var/export/pvs/${volume} 
    server: 192.168.0.254 
  persistentVolumeReclaimPolicy: Recycle 
EOF
     echo "Created def file for ${volume}"; \
done

Patch PVs definitions

for pv in $(oc get pv|awk '{print $1}' | grep pv | grep -v NAME); do oc patch pv $pv -p "spec:      
  accessModes:
  - ReadWriteMany
  - ReadWriteOnce
  - ReadOnlyMany
  persistentVolumeReclaimPolicy: Recycle"
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.