Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Don't run node config upgrade hook if host is not a node. | Devan Goodwin | 2016-08-23 | 1 | -1/+1 |
| | |||||
* | Merge pull request #2340 from lhuard1A/ansible_virt_module | Jason DeTiberus | 2016-08-22 | 5 | -41/+34 |
|\ | | | | | Replace some virsh commands by native virt_XXX ansible module | ||||
| * | Replace some virsh commands by native virt_XXX ansible module | Lénaïc Huard | 2016-08-22 | 5 | -41/+34 |
| | | |||||
* | | Merge pull request #2336 from sdodson/etcd-uninstall | Scott Dodson | 2016-08-22 | 1 | -1/+4 |
|\ \ | |/ |/| | Fix etcd uninstall | ||||
| * | Fix etcd uninstall | Scott Dodson | 2016-08-19 | 1 | -1/+4 |
| | | |||||
* | | Merge pull request #2330 from lhuard1A/open_node_ports | Jason DeTiberus | 2016-08-19 | 3 | -0/+14 |
|\ \ | | | | | | | Open OpenStack security group for the service node port range | ||||
| * | | Open OpenStack security group for the service node port range | Lénaïc Huard | 2016-08-19 | 3 | -0/+14 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | With OpenShift 3.2, creating a service accessible from the outside of the cluster thanks to `nodePort` automatically opens the “local” `iptables` firewall to allow incoming connection on the `nodePort` of the service. In order to benefit from this improvement, the OpenStack security group shouldn’t block those incoming connections. This change opens, on the OS nodes, the port range dedicated to service node ports. | ||||
* | | | Merge pull request #2329 from lhuard1A/fix_nodeonmaster | Jason DeTiberus | 2016-08-19 | 1 | -1/+1 |
|\ \ \ | |_|/ |/| | | Fix the “node on master” feature | ||||
| * | | Fix the “node on master” feature | Lénaïc Huard | 2016-08-19 | 1 | -1/+1 |
| |/ | | | | | | | | | | | | | | | | | | | | | | | | | What we want to do is to add the master as a node if: * `g_nodeonmaster` is set to true, and * we are not in the case where we want to add new nodes. The second test was done by only checking whether `g_new_node_hosts` was defined. This was wrong because, in all cloud-provider setups, this variable was set with the default value of “empty list” (`[]`). The test has been changed to use the `bool` filter so that it correctly evaluates to false (and hence, effectively add the master as a node) when `g_new_node_hosts` is the empty list. | ||||
* | | Merge pull request #2331 from dgoodwin/docker-upgrade-symlink | Scott Dodson | 2016-08-19 | 1 | -0/+1 |
|\ \ | | | | | | | Fix standalone Docker upgrade missing symlink. | ||||
| * | | Fix standalone Docker upgrade missing symlink. | Devan Goodwin | 2016-08-19 | 1 | -0/+1 |
| | | | |||||
* | | | Fix warnings in OpenStack provider with ansible 2.1 | Lénaïc Huard | 2016-08-19 | 3 | -22/+24 |
| |/ |/| | | | | | Some expressions now need to be enclosed inside `{{…}}`. | ||||
* | | Update uninstall.yml | Scott Dodson | 2016-08-17 | 1 | -1/+1 |
| | | | | | | Fixes #2317 | ||||
* | | Reconcile roles after master upgrade, but before nodes. | Devan Goodwin | 2016-08-16 | 1 | -43/+46 |
|/ | | | | | | | | | | Prevents the network egress bug causing node restart to fail during 3.3 upgrade. (even though a separate fix is incoming for this) Only catch is preventing the openshift_cli role, which requires docker, from triggering a potential upgrade, which we still don't want at this point. To avoid we use the same variable to protect docker installed version as we use in pre.yml. | ||||
* | Merge pull request #2220 from jkhelil/fix_key_error_when_ssh_failing | Scott Dodson | 2016-08-16 | 1 | -1/+1 |
|\ | | | | | fixing openshift key error in case of node failure during run (ssh is… | ||||
| * | fixing openshift key error in case of node failure during run (ssh issue) | jawed | 2016-08-09 | 1 | -1/+1 |
| | | |||||
* | | Reference tmpdir from first master hostvars when evacuating nodes. | Andrew Butcher | 2016-08-11 | 1 | -4/+4 |
| | | |||||
* | | Merge pull request #2285 from dgoodwin/nuke-images-2 | Scott Dodson | 2016-08-11 | 4 | -20/+59 |
|\ \ | | | | | | | Improvements for Docker 1.10+ Upgrade Image Nuking | ||||
| * | | Improvements for Docker 1.10+ upgrade image nuking. | Devan Goodwin | 2016-08-11 | 4 | -20/+55 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In a parallel step prior to real upgrade tasks, clear out all unused Docker images on all hosts. This should be relatively safe to interrupt as no real upgrade steps have taken place. Once into actual upgrade, we again clear all images only this time with force, and after stopping and removing all containers. Both rmi commands use a new and hopefully less error prone command to do the removal, this should avoid missed orphans as we were hitting before. Added some logging around the current image count before and after this step, most of them are only printed if we're crossing the 1.10 boundary but one does not, just for additional information in your ansible log. | ||||
| * | | Shutdown Docker before upgrading the rpm. | Devan Goodwin | 2016-08-10 | 1 | -0/+4 |
| |/ | | | | | | | | | This avoids the automatic image migration in 1.10, which can take a very long time and potentially cause rpm db corruption. | ||||
* / | Support for redeploying certificates. | Andrew Butcher | 2016-08-11 | 3 | -1/+270 |
|/ | |||||
* | Merge pull request #2211 from dgoodwin/33-upgrade-playbook | Scott Dodson | 2016-08-08 | 32 | -70/+227 |
|\ | | | | | 1.3 / 3.3 Upgrades | ||||
| * | Migrate ca.crt to ca-bundle.crt | Scott Dodson | 2016-08-08 | 1 | -0/+25 |
| | | |||||
| * | Upgrade configs for protobuf support. | Devan Goodwin | 2016-08-08 | 4 | -0/+68 |
| | | |||||
| * | Introduce 1.3/3.3 upgrade path. | Devan Goodwin | 2016-07-25 | 30 | -70/+134 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Refactored the 3.2 upgrade common files out to a path that does not indicate they are strictly for 3.2. 3.3 upgrade then becomes a relatively small copy of the byo entry point, all calling the same code as 3.2 upgrade. Thus far there are no known 3.3 specific upgrade tasks. In future we will likely want to allow hooks out to version specific pre/upgrade/post tasks. Also fixes a bug where the handlers were not restarting nodes/openvswitch containers doing upgrades, due to a change in Ansible 2+. | ||||
* | | Convert ansible facts callback to v2. | Andrew Butcher | 2016-08-08 | 1 | -1/+1 |
| | | |||||
* | | Call relocated openshift-loadbalancer playbook in master scaleup. | Andrew Butcher | 2016-08-04 | 1 | -0/+2 |
| | | |||||
* | | Merge pull request #2049 from abutcher/named-certificates | Scott Dodson | 2016-08-03 | 3 | -50/+12 |
|\ \ | | | | | | | Named CA Certificates | ||||
| * | | Add options for specifying named ca certificates to be added to the ↵ | Andrew Butcher | 2016-08-01 | 3 | -50/+12 |
| | | | | | | | | | | | | openshift ca bundle. | ||||
* | | | Refactor etcd certificates roles. | Andrew Butcher | 2016-08-02 | 3 | -269/+16 |
|/ / | |||||
* | | Move role dependencies to playbooks. | Andrew Butcher | 2016-07-29 | 3 | -3/+98 |
| | | |||||
* | | Merge pull request #2236 from maxamillion/fedora | Scott Dodson | 2016-07-29 | 1 | -1/+1 |
|\ \ | | | | | | | update bootstrap-fedora playbook with new python crypto deps | ||||
| * | | update bootstrap-fedora playbook with new python crypto deps | Adam Miller | 2016-07-28 | 1 | -1/+1 |
| | | | | | | | | | | | | Signed-off-by: Adam Miller <maxamillion@fedoraproject.org> | ||||
* | | | Merge pull request #2210 from abutcher/staggered-master-start | Scott Dodson | 2016-07-28 | 1 | -0/+1 |
|\ \ \ | |/ / |/| | | Stagger HA master service start | ||||
| * | | Stagger the start of master services. | Andrew Butcher | 2016-07-24 | 1 | -0/+1 |
| | | | |||||
* | | | Merge pull request #2213 from detiber/remove_old_metrics_role | Jason DeTiberus | 2016-07-26 | 1 | -2/+0 |
|\ \ \ | | | | | | | | | remove outdated openshift_cluster_metrics role | ||||
| * | | | remove outdated openshift_cluster_metrics role | Jason DeTiberus | 2016-07-25 | 1 | -2/+0 |
| | | | | |||||
* | | | | Template named certificates with_items. | Andrew Butcher | 2016-07-25 | 1 | -2/+2 |
| | | | | |||||
* | | | | Replace master_cert_config_dir with common config_base fact. | Andrew Butcher | 2016-07-25 | 1 | -1/+1 |
|/ / / | |||||
* | | | Merge pull request #2196 from dgoodwin/nuke-images-symlink | Andrew Butcher | 2016-07-22 | 1 | -0/+1 |
|\ \ \ | |/ / |/| | | Add missing nuke_images.sh symlink. | ||||
| * | | Add missing nuke_images.sh symlink. | Devan Goodwin | 2016-07-22 | 1 | -0/+1 |
| |/ | |||||
* | | Merge pull request #1990 from abutcher/openshift-certificates | Scott Dodson | 2016-07-21 | 2 | -154/+16 |
|\ \ | |/ |/| | Refactor openshift certificates roles. | ||||
| * | Refactor openshift certificates roles. | Andrew Butcher | 2016-07-20 | 2 | -154/+16 |
| | | |||||
* | | Merge pull request #2168 from dgoodwin/container-cli-speed | Scott Dodson | 2016-07-21 | 1 | -0/+2 |
|\ \ | | | | | | | Copy openshift binary instead of using wrapper script. | ||||
| * | | Stop reporting changes when docker pull is already up to date. | Devan Goodwin | 2016-07-20 | 1 | -0/+2 |
| | | | |||||
* | | | Merge pull request #2187 from lhuard1A/bin_cluster_libvirt_ansible_2.1 | Jason DeTiberus | 2016-07-21 | 4 | -21/+24 |
|\ \ \ | | | | | | | | | Fix libvirt provider for Ansible 2.1.0.0 | ||||
| * | | | Fix libvirt provider for Ansible 2.1.0.0 | Lénaïc Huard | 2016-07-21 | 4 | -21/+24 |
| | |/ | |/| | |||||
* | | | Correct relative include for ansible version check. | Andrew Butcher | 2016-07-21 | 1 | -1/+1 |
| | | | |||||
* | | | Check ansible version prior to evaluating cluster hosts and groups. | Andrew Butcher | 2016-07-20 | 7 | -0/+24 |
|/ / | |||||
* | | Merge pull request #2175 from dgoodwin/origin-12-rpm-upgrade | Scott Dodson | 2016-07-19 | 1 | -2/+6 |
|\ \ | | | | | | | Fix bugs with origin 1.2 rpm based upgrades. |