Releases: lae/ansible-role-proxmox
v1.6.3
Bug fix release for PVE 6.x, and HA group configuration support
Changes/Fixes
#50 - HA groups can now be managed
#80 - PVE auth users are now properly created with specified passwords (and other attributes)
#81 - Idempotency checks for Ceph OSD creation is now more correct
#87 - Size attributes can now be defined for Ceph pools
#88 - Subscription patch no longer fails on PVE 6.x
a3720a9 - Vagrant deployment also brings up a Ceph storage cluster
v1.6.2
Minor bug fix release and Vagrantfile addition.
#76 Vagrant can now be used to bring up a local PVE cluster. It's also now possible to use Python 3 as the interpreter when deploying this role.
#75 ring0/link0 and related role variables have been replaced with pve_cluster_addr0/1
#74 pve_ceph_cluster_network can now be specified to use another network for backend traffic
#70 apt dist-upgrade is now used in lieu of apt upgrade when pve_run_system_upgrades is true
#69 Ceph improvements, namely one to help idempotency. (Ceph testers needed!)
DEPRECATION NOTICE
Support for the following role variables will be removed in the next minor version (1.7.0):
pve_cluster_ring0_addr
pve_cluster_ring1_addr
pve_cluster_bindnet0_addr
pve_cluster_bindnet1_addr
pve_cluster_link0_addr
pve_cluster_link1_addr
If you are overriding any of these, please update your playbooks to use pve_cluster_addr0 and/or pve_cluster_addr1 before updating to v1.7.0.
v1.6.1
Bug fix release for Ceph deployments on PVE 6/Debian Buster.
#63 - Introduces pve_ceph_repository_line for customizing the Ceph package repository used (and defaults to the correct one per version)
#64 - Introduces a pve_ssh_port that allows you to specify which SSH port that hosts in a cluster are listening on. This does not reconfigure SSH to listen on this port; rather, it configures the SSH client what port to use when connecting to other hosts in the cluster.
v1.6.0
v1.5.0
v1.4.12
This is an important bug fix release. Please upgrade to this release to prevent issues adding nodes to a cluster or creating a new cluster.
42495cc Fixes add node tasks during cluster creation for PVE 5.x
0afc625 Preliminary support for Ansible 2.6 (current deprecation warnings resolved)
b010844 Testing environment refresh, with addition of tests for PVE 4.x on older Ansible versions
v1.4.11
This feature release includes the following:
#28 - Let's Encrypt SSL certificate support (see README for details)
#27 - apt-get update and apt-get dist-upgrade are no longer run by default on every run.
#26 - Limit root login between PVE cluster hosts to key authentication
The pve_upgrade variable has been deprecated in this release for the solution in #27. If you wish to use this role to keep installing system updates, please specify pve_run_system_upgrades: true in your playbook.
v1.4.10
Minor feature release.
By default, this role executes a apt-get dist-upgrade on every run. While this is required for initial installation, some users may not want this to occur when re-running their playbook after an initial installation run. You can configure this with the pve_upgrade role variable: no will disable all upgrades and safe will execute aptitude safe-upgrade (see apt_module docs for more information).
#25 - Allow upgrade step to be configurable
v1.4.9
Bug fix release for the subscription removal patch, as well as fixed a regression from 1.4.8 that removed enterprise repos for enterprise repo users.
v1.4.8
This is a quick feature release in order to allow users to easily disable the subscription message removal patches using the newly introduced role variable pve_remove_subscription_warning before #24 is properly fixed.
When running your playbook, please pass -e '{"pve_remove_subscription_warning": false}' to ansible-playbook in order to skip the failing task (you can also put it in your host/group variables if you wish if you want to make it permanent). All this will do is skip patching the Proxmox user interface to remove the subscription message if you are not using the enterprise repository.