| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
firewall changes have been applied.
|
| |
|
|\
| |
| | |
Support openshift_node_port_range for configuring service NodePorts
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Sets the appropriate config field if openshift_node_port_range is set
and also configures filewalls on each node. firewalld already supports
port ranges like "30000-32000", while iptables needs that value
converted to the correct "30000:32000" form for use with `--dport`.
If not set, no node ports are opened.
|
|/
|
|
| |
Closes #3070
|
|
|
|
| |
and openshift_hosted.
|
|
|
|
| |
* https://trello.com/c/TeaEB9fX/307-3-deprecate-node-evacuation
|
|
|
|
|
| |
* Added checks to make ci for yaml linting
* Modified y(a)ml files to pass lint checks
|
|\
| |
| | |
Pre-pull master/node/ovs images during upgrade.
|
| |
| |
| |
| |
| |
| |
| |
| | |
We did this for install but not upgrade, leading to situations where the
service restarts after upgrade could take much longer than expected as
docker pulls down the new image. Now the images are present when we
restart services and should allow them to come back online much more
quickly, equivalent to rpm service restarts.
|
|/ |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In 3.3 one of our services lays down a systemd drop-in for configuring
Docker networking to use lbr0. In 3.4, this has been changed but the
file must be cleaned up manually by us.
However, after removing the file docker requires a restart. This had big
implications particularly in containerized environments where upgrade is
a very fragile series of upgrading and service restarts.
To avoid double docker restarts, and thus double service restarts in
containerized environments, this change does the following:
- Skip restart during docker upgrade, if it is required. We will restart
on our own later.
- Skip containerized service restarts when we upgrade the services
themselves.
- Clean shutdown of all containerized services.
- Restart Docker. (always, previously this only happened if it needed an
upgrade)
- Ensure all containerized services are restarted.
- Restart rpm node services. (always)
- Mark node schedulable again.
At the end of this process, docker0 should be back on the system.
|
|
|
|
|
|
| |
* Ansible systemd module used in place of service module
* Refactored command tasks which are no longer necessary
* Applying rules from openshift-ansible Best Practices Guide
|
|\
| |
| | |
Refactor to use Ansible package module
|
| |
| |
| |
| |
| | |
The Ansible package module will call the correct package manager for the
underlying OS.
|
|/
|
|
|
|
| |
Fedora Atomic Host does not have tuned installed.
Fixes #2809
|
|\
| |
| | |
Support 3rd party scheduler
|
| | |
|
|\ \
| | |
| | | |
[#2698] Change to allow cni deployments without openshift SDN
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The roles/openshift_facts main task did not pass the cni plugin variable to the later role playbooks.
The master.yaml and node.yaml templates did not allow for a cni configuration without either installing openshift sdn or nuage.
This change will allow to use os_sdn_network_plugin_name=cni and set openshift_use_openshift_sdn=false for deployments that use a cni plugin that doesn't need and want openshift sdn to be installed
|
|/ / |
|
| |
| |
| |
| | |
This reverts commit 1f2276fff1e41c1d9440ee8b589042ee249b95d7.
|
| | |
|
| |
| |
| |
| |
| |
| | |
At recommendation of network team, more details in:
https://bugzilla.redhat.com/show_bug.cgi?id=1331590
|
|/
|
|
| |
* Node labels are parsed from openshift_node_labels if they exist
|
| |
|
| |
|
|\
| |
| | |
nfs: Handle seboolean aliases not just in Fedora
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
I'm testing with a bleeding edge RHEL Atomic Host, and it looks
like we pulled in a new version of selinux-policy that has
`virt_sandbox_use_nfs` aliased to `virt_use_nfs`.
In https://github.com/openshift/openshift-ansible/pull/2356
Adam changed this to check for Fedora. This changes things
to drop the distribution check, and instead parse the `getsebool`
output to determine whether or not the boolean is an alias,
and should hence work on all distributions/versions.
|
|\|
| |
| |
| | |
variable_with_no_default_2576
|
| | |
|
|/ |
|
| |
|
|\
| |
| | |
Fix openstack cloudprovider
|
| | |
|
| | |
|
|\ \
| | |
| | | |
Add portal_net and sdn_cluster_network_cidr to node NO_PROXY
|
| |/ |
|
|\ \
| |/
|/| |
Add atomic-guest tuned profile
|
| | |
|
| | |
|
| | |
|
|/
|
|
|
|
|
|
| |
The openvswitch restart appears to hang the node container running
briefly, if you try to restart node too quickly it will fail complaining
that the node container name is still in use. Pausing after openvswitch
restart allows the existing node container to catch up and then shutdown
cleanly.
|
|
|
|
|
| |
Reference:
https://bugzilla.redhat.com/show_bug.cgi?id=1372388
|
| |
|
|\
| |
| | |
Reload units after node container service modified.
|
| | |
|
|/ |
|
|
|
|
| |
Signed-off-by: Adam Miller <maxamillion@fedoraproject.org>
|
|
|
|
| |
Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1367937
|