Installing 3scale AMP 2.2 on OCP 3.9 on ec2
Note: This is not a supported configuration. For PoC, recommended OCP setup is 1 master 3 node. `oc cluster up` is good way to get 3scale AMP up an running quickly.
You will need:
-
a running instance with 8GB RAM minimum (recommended 16GB) and RHEL
-
<PUBLIC_DNS>: (e.g.ec2-54-123-456-78.compute-1.amazonaws.com) -
<PUBLIC_IP>: (e.g.54.123.456.78)
Set up OpenShift cluster
References
Install and run Docker
sudo yum-config-manager --enable rhui-REGION-rhel-server-extras
sudo yum install docker docker-registry -y
/etc/containers/registries.conf:
registries = ['172.30.0.0/16']
sudo systemctl start docker
sudo systemctl status docker
Install OC tools
OCP: https://mirror.openshift.com/pub/openshift-v3/clients/3.9.14/linux/oc.tar.gz
Example:
sudo yum install wget -y
wget https://mirror.openshift.com/pub/openshift-v3/clients/3.9.14/linux/oc.tar.gz
tar xzvf oc.tar.gz
mv oc /usr/bin/
rm -rf oc.tar.gz
Start the cluster
oc cluster up --public-hostname=<PUBLIC_DNS> --routing-suffix=<PUBLIC_IP>.xip.io --host-data-dir=/home/ec2-user/myoccluster1
Check out the console:
https://<PUBLIC_DNS>:8443
Deploy 3scale AMP
Create persistent volumes
sudo su
mkdir -p /var/lib/docker/pv/{01..04}
chmod g+w /var/lib/docker/pv/{01..04}
chcon -Rt svirt_sandbox_file_t /var/lib/docker/pv/
(pv.yml - copy from resources folder)
oc login -u system:admin
oc new-app --param PV=01 -f pv.yml
oc new-app --param PV=02 -f pv.yml
oc new-app --param PV=03 -f pv.yml
oc new-app --param PV=04 -f pv.yml
oc get pv
(amp.yml - copy from resources folder)
Login as developer and start AMP with template
oc login https://<PUBLIC_DNS>:8443 --insecure-skip-tls-verify
oc new-project 3scale-amp
oc new-app --file amp.yml --param TENANT_NAME=3scale --param WILDCARD_DOMAIN=<PUBLIC_IP>.xip.io >> /tmp/3scale_amp_provision_details.txt
cat /tmp/3scale_amp_provision_details.txt
--> Deploying template "3scale-amp/system" for "amp.yml" to project 3scale-amp
system
---------
Login on Login on https://3scale-admin.54.86.18.216.xip.io as admin/gu8edykg <===== LOGIN with these credentials
* With parameters:
* AMP_RELEASE=er3
* ADMIN_PASSWORD=gu8edykg # generated
* ADMIN_USERNAME=admin
* APICAST_ACCESS_TOKEN=rthdeuql # generated
* ADMIN_ACCESS_TOKEN=4o2txf0v4e3wgvtw # generated
* WILDCARD_DOMAIN=<PUBLIC_IP>.xip.io
* SUBDOMAIN=3scale
* MySQL User=mysql
* MySQL Password=qfnt75jf # generated
* MySQL Database Name=system
* MySQL Root password.=7dhquse7 # generated
* SYSTEM_BACKEND_USERNAME=3scale_api_user
* SYSTEM_BACKEND_PASSWORD=a3i3n7by # generated
* REDIS_IMAGE=rhscl/redis-32-rhel7:3.2-5.3
* SYSTEM_BACKEND_SHARED_SECRET=s4wpndxj # generated
RESUME PODS
1. Resume database tier pods
for x in backend-redis system-memcache system-mysql system-redis zync-database; do echo Resuming dc: $x; sleep 2; oc rollout resume dc/$x; done
Verify all pods are running
oc get pods
NAME READY STATUS RESTARTS AGE
backend-redis-1-q2hnc 1/1 Running 0 53s
system-memcache-1-ggjgm 1/1 Running 0 49s
system-mysql-1-pg7rm 1/1 Running 0 1m
system-redis-1-klthg 1/1 Running 0 44s
zync-database-1-w66qf 1/1 Running 0 52s
2. Resume backend listener and worker deployments:
for x in backend-listener backend-worker; do echo Resuming dc: $x; sleep 2; oc rollout resume dc/$x; done
3. Resume the system-app and its two containers:
oc rollout resume dc/system-app
4. Resume additional system and backend application utilities:
for x in system-resque system-sidekiq backend-cron system-sphinx; do echo Resuming dc: $x; sleep 2; oc rollout resume dc/$x; done
5. Resume apicast gateway deployments:
for x in apicast-staging apicast-production; do echo Resuming dc: $x; sleep 2; oc rollout resume dc/$x; done
6. Resume remaining deployments:
for x in apicast-wildcard-router zync; do echo Resuming dc: $x; sleep 2; oc rollout resume dc/$x; done