search

OpenShiftDemos / openshift-ops-workshops

Workshop materials for OpenShift admin training, covering Red Hat OpenShift Container Platform 4, Modern App Dev Roadshow's Ops Track, and Summit 2023 Hands-on with OCP Plus.
GNU General Public License v3.0
118 stars 151 forks source link

Playbook for installing OCP stops everytime at the task - 'Start and enable iptables service' #30

Closed kmurudi closed 7 years ago

kmurudi commented 7 years ago

using this command to run installer- ansible-playbook /usr/share/ansible/openshift-ansible/playbooks/byo/config.yml

always stops at the mentioned task and then have to stop and run the playbook again..gets deployed second time though.

thoraxe commented 7 years ago

Is this still a problem now that we have made the machines bigger?

kmurudi commented 7 years ago

can start installing after 2-3 minutes after ssh-ing into master host. will test again and let know by evening

thoraxe commented 7 years ago

Any updates? IDM issues are separate.

dmesser commented 7 years ago

I think the workaround Ryan introduced to pre-enable iptables.service and disable firewalld.service works. You just have to wait for cloud-init to finish and ensure that iptables.service is indeed active. Then you can start the install and it works. With the previous cloud-init steps this (iptables active) could take 2-3 minutes to be done due to long-running bash scripts. Now with cloud-init executing ansible playbooks it was much faster for me (~40-60 secs after instance boot) last time I checked. Eventually we can use a wait-condition to signal the user successful rollout of the lab from a CFN standpoint so that the wait-condition would only fire when the last step of cloud-init ran successfully. That would include iptables.service activation and the user only enters the lab environment when this is definitely done and can rollout out OpenShift immediately.