| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
| |
|
|\
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Automatic merge from submit-queue.
Fix version multi
Simplify double upgrade version logic
Currently, double upgrade process (3.7 -> 3.9)
for control plane attempts to run openshift_version
role twice to set the appropriate values for
upgrading each major version, 3.8 and 3.9.
This commit instructs openshift_version to
only inquire about the proper settings for 3.9,
and hard-sets the appropriate values for 3.8.
This allows a simplification of the
openshift_version role, allowing for easier
debugging.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1542368
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Currently, double upgrade process (3.7 -> 3.9)
for control plane attempts to run openshift_version
role twice to set the appropriate values for
upgrading each major version, 3.8 and 3.9.
This commit instructs openshift_version to
only inquire about the proper settings for 3.9,
and hard-sets the appropriate values for 3.8.
This allows a simplification of the
openshift_version role, allowing for easier
debugging.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1542368
|
|\ \
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Automatic merge from submit-queue.
Fix docker_upgrade variable
Currently, docker_upgrade is ignored during
cluster upgrades.
This commit ensures that the variable is respected.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1543714
|
| |/
| |
| |
| |
| |
| |
| |
| |
| | |
Currently, docker_upgrade is ignored during
cluster upgrades.
This commit ensures that the variable is respected.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1543714
|
|\ \
| | |
| | | |
Invert logic to decide when to re-deploy certs
|
| |/ |
|
|\ \
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Automatic merge from submit-queue.
Add service catalog to be upgraded
Currently service catalog was left running at the previous version.
https://bugzilla.redhat.com/show_bug.cgi?id=1541247
|
| |/
| |
| |
| | |
Currently service catalog was left running at the previous version.
|
|/ |
|
|
|
|
|
|
|
|
|
|
|
| |
An entry-point playbook was imported by mistake.
This caused common init code to run again, which
is undesireable.
This commit changes the import to use the corresponding
'private' play which does not call the init code.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1542855
|
|\
| |
| |
| |
| |
| |
| | |
Automatic merge from submit-queue.
Redeploy router certificates during upgrade only when secure.
Wrap the upgrade logic for redeploying certificates into another block so that insecure registries do not perform any certificate tasks.
|
| | |
|
|\ \
| |/
|/| |
Determine which host is the etcd CA host
|
| | |
|
|\ \
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Automatic merge from submit-queue.
[1540537] Add base package installation to upgrade playbooks
Hosts will need python ipaddress module installed if it was not
installed during initial installation.
Bug 1540537
https://bugzilla.redhat.com/show_bug.cgi?id=1540537
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Hosts will need python ipaddress module installed if it was not
installed during initial installation.
Bug 1540537
https://bugzilla.redhat.com/show_bug.cgi?id=1540537
|
|\ \ \
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Automatic merge from submit-queue.
3.9 upgrade: fix typos in restart masters procedure
* 'rolling_restart_mode' should be 'services', not 'service'
* use 'state: restarted' to properly restart services
Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1540054
Signed-off-by: Vadim Rutkovsky <vrutkovs@redhat.com>
|
| |/ /
| | |
| | |
| | |
| | |
| | |
| | | |
* 'rolling_restart_mode' should be 'services', not 'service'
* use 'state: restarted' to properly restart services
Signed-off-by: Vadim Rutkovsky <vrutkovs@redhat.com>
|
|/ / |
|
|\ \
| | |
| | | |
Initial support for 3.10
|
| | | |
|
|\ \ \
| | | |
| | | | |
Upgrades: pass openshift_manage_node_is_master to master nodes during upgrade
|
| |/ /
| | |
| | |
| | | |
This ensures required labels for master would be set
|
| | | |
|
|/ / |
|
|\ \
| | |
| | | |
Fix version handling in 3.8/3.9 control plane upgrades
|
| | | |
|
| | | |
|
|\ \ \
| |/ /
|/| |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Automatic merge from submit-queue.
Upgrade playbook improvements
Two changes in this PR:
* Avoid duplication in upgrade.yml - it now would simply run upgrade_control_plane and upgrade_nodes playbooks
* Restart controllers instead of start/stop if restart mode is 'services'
* For 'system' restart mode controllers are disabled before master upgrade and re-enabled after upgrades are finished
|
| | |
| | |
| | |
| | |
| | |
| | | |
mode is system
This removes a start/stop after control plane upgrade
|
| | |
| | |
| | |
| | | |
upgrade
|
| | | |
|
|\ \ \
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Automatic merge from submit-queue.
upgrades: use openshift_version as a regexp when checking openshift.common.version
This would fix errors like 'Master running 3.9.0 must be upgraded to
3.9* before node upgrade can be run' during the upgrade
Signed-off-by: Vadim Rutkovsky <vrutkovs@redhat.com>
|
| |/ /
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
openshift.common.version
This would fix errors like 'Master running 3.9.0 must be upgraded to
3.9* before node upgrade can be run' during the upgrade
Signed-off-by: Vadim Rutkovsky <vrutkovs@redhat.com>
|
|\ \ \
| |_|/
|/| | |
Remove master_ha bool checks
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Most of these checks are no longer applicable to
the tasks on which they are applied.
This commit removes incorrect ha checks to ensure
services are restarted at appropriate times.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1500897
|
| |/
|/|
| |
| |
| |
| |
| |
| |
| |
| | |
Version 3.8 is an intermediate step when upgrading to 3.9 and
openshift_pkg_version may be set in the inventory for 3.9. This will
ensure there is not a conflict between the intermediate step and the
final version requested.
Bug 1540108
https://bugzilla.redhat.com/show_bug.cgi?id=1540108
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Due to complexities upgrading two versions
at the same time, openshift_image_tag was being
set incorrectly during control_plane upgrades.
This commit ensures that openshift_image_tag
is set correctly during this process.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1536839
|
|\ \
| | |
| | | |
Remove old assetConfig from master-config.yaml
|
| |/
| |
| |
| |
| |
| |
| |
| | |
The `assetConfig` is no longer used now that the console is split into
its own pod.
- Remove `assetConfig` on upgrades to 3.9
- Stop writing logging and metrics URLs to assetConfig for 3.9
|
|\ \
| |/
|/|
| |
| | |
Automatic merge from submit-queue.
Bug 1536217: Need to validate etcd serving certs before 3.9 upgrade
|
| |
| |
| |
| | |
etcd serving cert SAN.
|
|\ \
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Automatic merge from submit-queue.
Label masters with node-role.kubernetes.io/master
This is controlled by `openshift_master_node_label` var in openshift_manage_node
Related to https://bugzilla.redhat.com/show_bug.cgi?id=1535673
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
labels
and scheduling status during upgrades
Signed-off-by: Vadim Rutkovsky <vrutkovs@redhat.com>
|
|\ \ \
| | | |
| | | |
| | | |
| | | | |
mgugino-upstream-stage/protect-upgrade-master-version
Protect master installed version during node upgrades
|
| |/ /
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Master version is unprotected during all upgrades.
During node-only upgrades, master version should
remain protected to ensure correct image tag is
set for node upgrades.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1536839
|
|\ \ \
| |/ /
|/| | |
Allow openshift-ansible image to deploy to GCP
|
| | | |
|