[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: Migration from 3.10 to 3.11

* Samuel Martín Moro <faust64 gmail com> [2019-04-17 15:46:00 +0200] wrote :


> Hi,


> Deploying or upgrading to 3.11, I'ld go with ansible 2.6.
> Currently, the ansible package provided by RedHat is some 2.6.16-1.
> Personally, I'm still using 2.6.5, from PIP. I'm not sure 2.7 is actually
> supported.

FYI, I'm using Centos and installed ansible 2.6.5 as you suggested even
if the README indicates that 2.7 is supported.

> To upgrade an existing installation, one would usually:
> - check the release notes (
> https://docs.openshift.com/container-platform/3.11/release_notes/ocp_3_11_release_notes.html
> )
> - update your inventory accordingly (remove deprecated variables, add what
> you'ld need, ...)
> - follow the in-place cluster upgrades procedure (
> https://docs.openshift.com/container-platform/3.11/upgrading/automated_upgrades.html
> )

Actually, I checked that link :


which looks quite the same with less stuff to do apparently.

> openshift_image_tag: v3.11
> openshift_pkg_version: -3.11.0
> openshift_release: v3.11
> openshift_upgrade_min: 3.11
> openshift_version: "3.11"

Done this. Wondering if that's a typo error for : openshift_pkg_version:

> Once your variables are ready, you should be able to apply the
> upgrade_control_plane and upgrade_nodes playbooks:
> ansible-playbook -i </path/to/inventory/file>
> ./playbooks/byo/openshift-cluster/upgrades/v3_11/upgrade_control_plane.yml
> ansible-playbook -i </path/to/inventory/file>
> ./playbooks/byo/openshift-cluster/upgrades/v3_11/upgrade_nodes.yml

Shouldn't it be
playbooks/byo/openshift-cluster/upgrades/v3_11/upgrade.yml ?

Of what I understand it does backups and some other tasks.


> Note that upgrading, you shouldn't have to apply the prerequisites
> playbook. Yet it should work.
> If you can't get it to work, then try forcing IMAGE_VERSION=v3.10 in
> /etc/sysconfig/origin-node. Let us know how that goes.

Thte version variable was already in the file on each master.

Don't have that error anymore with this version of ansible but got that
one :

     Play:     Install packages necessary for installer
     Task:     Install ntp package
     Message:  The conditional check 'chrony_installed.rc != 0' failed. The error was: error while evaluating conditional (chrony_installed.rc != 0): 'dict object' has no attribute 'rc'

               The error appears to have been in 'openshift-ansible/playbooks/init/base_packages.yml': line 16, column 5, but may
               be elsewhere in the file depending on the exact syntax problem.

               The offending line appears to be:

                 # atomic here.
                 - name: Install ntp package

But ntp and chrony are both installed on each node.

Thanks for the tips,

Kind regards,

Hyde Stevenson

Consola-te com suportar as injustiças: a verdadeira desgraça está em
                -- Pitágoras

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]