Description of problem: when the OCP v3.6 was upgraded from v3.5 ,the registry dns wasn't enabled by default. But if we scale up master against this OCP 3.6 env, the registry dns was enabled on the added master. The new added master configuration should use same with the existing master. Version-Release number of the following components: openshift-ansible-3.6.140 How reproducible: always Steps to Reproduce: 1. install HA OCP v3.5 2. upgrade to OCP 3.6 anible-playbook /usr/share/ansible/openshift-ansible/playbooks/byo/openshift-cluster/upgrades/v3_6/upgrade.yml 3. Add a new master. ansible-playbook --private-key /root/configdir/libra.pem -i /root/workdir/hosts /usr/share/ansible/openshift-ansible/playbooks/byo/openshift-master/scaleup.yml -v 4. check the registry dns configuration cat /etc/sysconfig/atomic-openshift-master-api Actual results: 4. There is OPENSHIFT_DEFAULT_REGISTRY=docker-registry.default.svc:5000 for the added master [4]. These aren't OPENSHIFT_DEFAULT_REGISTRY for the upgraded masters [1][2][3]. [1]openshift-221.lab.eng.nay.redhat.com | SUCCESS | rc=0 >> OPTIONS=--loglevel=2 --listen=https://0.0.0.0:8443 --master=https://openshift-221.lab.eng.nay.redhat.com:8443 CONFIG_FILE=/etc/origin/master/master-config.yaml IMAGE_VERSION=v3.6.126.8 # Proxy configuration # See https://docs.openshift.com/enterprise/latest/install_config/install/advanced_install.html#configuring-global-proxy [2]openshift-182.lab.eng.nay.redhat.com | SUCCESS | rc=0 >> OPTIONS=--loglevel=2 --listen=https://0.0.0.0:8443 --master=https://openshift-182.lab.eng.nay.redhat.com:8443 CONFIG_FILE=/etc/origin/master/master-config.yaml IMAGE_VERSION=v3.6.126.8 # Proxy configuration # See https://docs.openshift.com/enterprise/latest/install_config/install/advanced_install.html#configuring-global-proxy [3]openshift-181.lab.eng.nay.redhat.com | SUCCESS | rc=0 >> OPTIONS=--loglevel=2 --listen=https://0.0.0.0:8443 --master=https://openshift-181.lab.eng.nay.redhat.com:8443 CONFIG_FILE=/etc/origin/master/master-config.yaml IMAGE_VERSION=v3.6.126.8 # Proxy configuration # See https://docs.openshift.com/enterprise/latest/install_config/install/advanced_install.html#configuring-global-proxy [4]openshift-210.lab.eng.nay.redhat.com | SUCCESS | rc=0 >> OPTIONS=--loglevel=2 --listen=https://0.0.0.0:8443 --master=https://openshift-210.lab.eng.nay.redhat.com:8443 CONFIG_FILE=/etc/origin/master/master-config.yaml OPENSHIFT_DEFAULT_REGISTRY=docker-registry.default.svc:5000 IMAGE_VERSION=v3.6.126.8 # Proxy configuration # See https://docs.openshift.com/enterprise/latest/install_config/install/advanced_install.html#configuring-global-proxy NO_PROXY=,172.30.0.0/16,10.128.0.0/14 Expected results: The new added master has same configuration as the existing masters Additional info:
Created attachment 1296020 [details] The scaleup logs
Successfully reproduced this error
PR for this is open here https://github.com/openshift/openshift-ansible/pull/4894 Need to clean out some debugging trash and squash the commits down.
Doing a last batch of testing. Attempting to reproduce the original bug. If I can not reproduce the bug then I'll switch the github portion of this over to requesting reviews so we can get the fix merged.
With openshift3/ose-ansible:v3.6.173.0.27, the scale master is still include OPENSHIFT_DEFAULT_REGISTRY. [root@anli host3ha]# sshremote masters 'cat /etc/sysconfig/atomic-openshift-master-api |grep -v "#"' openshift-182.lab.eng.nay.redhat.com | SUCCESS | rc=0 >> OPTIONS=--loglevel=2 --listen=https://0.0.0.0:8443 --master=https://openshift-182.lab.eng.nay.redhat.com:8443 CONFIG_FILE=/etc/origin/master/master-config.yaml openshift-181.lab.eng.nay.redhat.com | SUCCESS | rc=0 >> OPTIONS=--loglevel=2 --listen=https://0.0.0.0:8443 --master=https://openshift-181.lab.eng.nay.redhat.com:8443 CONFIG_FILE=/etc/origin/master/master-config.yaml openshift-221.lab.eng.nay.redhat.com | SUCCESS | rc=0 >> OPTIONS=--loglevel=2 --listen=https://0.0.0.0:8443 --master=https://openshift-221.lab.eng.nay.redhat.com:8443 CONFIG_FILE=/etc/origin/master/master-config.yaml openshift-210.lab.eng.nay.redhat.com | SUCCESS | rc=0 >> OPTIONS=--loglevel=2 --listen=https://0.0.0.0:8443 --master=https://openshift-210.lab.eng.nay.redhat.com:8443 CONFIG_FILE=/etc/origin/master/master-config.yaml OPENSHIFT_DEFAULT_REGISTRY=docker-registry.default.svc:5000
Created attachment 1320589 [details] scaleup logs
I filed a 3.6 backport for this fix here https://github.com/openshift/openshift-ansible/pull/5952
should now be fixed based on comment 10
Verified in openshift-ansible-3.6.173.0.101-1.git.0.7c6b77f.el7.noarch.rpm OPENSHIFT_DEFAULT_REGISTRY not set as the fist master after scaling up masters.