summaryrefslogtreecommitdiffstats
path: root/roles
Commit message (Collapse)AuthorAgeFilesLines
* cri-o: Restart cri-o after openshift sdn installationSteve Milner2017-08-031-0/+7
|
* cri-o: Continue node without SELinux checkSteve Milner2017-08-031-3/+3
| | | | | cri-o currently requires SELinux to be off. This change disables the SELinux check in the openshift_node role when cri-o is in use.
* cri-o: allow to override CRI-O image indipendently from DockerGiuseppe Scrivano2017-08-031-3/+3
| | | | Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
* docker: introduce use_crio_onlyGiuseppe Scrivano2017-08-031-2/+3
| | | | | | | | Introduce a new variable that disable the installation of Docker. For the time being we will still need Docker for building images, so by default leave it installed. Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
* docker: skip Docker setup when using CRI-OGiuseppe Scrivano2017-08-031-1/+3
| | | | Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
* openvswitch: system container depends on the cri-o serviceGiuseppe Scrivano2017-08-031-1/+12
| | | | Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
* cli_image: do not require Docker when using CRI-OGiuseppe Scrivano2017-08-032-25/+83
| | | | | | Use atomic to copy the CLI binaries to the host. Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
* cri-o: skip Set precise containerized version checkGiuseppe Scrivano2017-08-031-0/+9
| | | | Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
* cri-o: skip Docker version testGiuseppe Scrivano2017-08-031-0/+2
| | | | Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
* cri-o: use only images from Docker HubGiuseppe Scrivano2017-08-031-10/+6
| | | | | | | For the time being it won't be added to the Red Hat registry, so use only what is available on Docker Hub. Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
* cri-o: Enable systemd-modules-load if requiredSteve Milner2017-08-031-0/+6
| | | | | If we had to drop the overlay file in /etc/modules-load.d/ then enable the systemd-modules-load service and make sure it runs.
* openshift_node: fix typo for experimental-criSteve Milner2017-08-031-1/+1
|
* cri-o: Fix node template to use full variableSteve Milner2017-08-031-1/+1
|
* cri-o: Ensure overlay is availableSteve Milner2017-08-032-0/+22
| | | | | | | Some distro releases may not have overlay loaded into the kernel. This change looks for overlay via lsmod and, if it isn't already there, uses modprobe to load it in and then drops a load config into /etc/modules-load.d/overlay.conf.
* cri-o: Default insecure registries to ""Steve Milner2017-08-032-1/+2
|
* crio: use a template for the configurationGiuseppe Scrivano2017-08-032-24/+137
| | | | Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
* openshift_docker_facts: Add use_crioSteve Milner2017-08-031-0/+1
|
* cri-o: Minor fixes for tasksSteve Milner2017-08-031-2/+1
|
* cri-o: Hardcode image name to cri-oSteve Milner2017-08-031-1/+1
|
* cri-o: Add cri-o as a Wants in node unitsSteve Milner2017-08-032-1/+2
|
* cri-o: configure storage and insecure registriesGiuseppe Scrivano2017-08-032-3/+34
| | | | Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
* node.yaml: configure node to use cri-o when openshift.common.use_crioGiuseppe Scrivano2017-08-031-0/+15
| | | | Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
* cri-o: Allow cri-o usage.Steve Milner2017-08-032-0/+110
| | | | This change reuses the docker role to inject cri-o usage.
* Merge pull request #3043 from jkhelil/scaleup_etcdScott Dodson2017-08-023-11/+9
|\ | | | | Add etcd scaleup playbook
| * add etcd scaleup playbookjawed2017-06-153-11/+9
| |
* | Merge pull request #4981 from jarrpa/heketi-key-fixScott Dodson2017-08-021-1/+6
|\ \ | | | | | | GlusterFS: Copy SSH private key to master node.
| * | GlusterFS: Copy SSH private key to master node.Jose A. Rivera2017-08-021-1/+6
| | | | | | | | | | | | | | | | | | Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1477718 Signed-off-by: Jose A. Rivera <jarrpa@redhat.com>
* | | Merge pull request #4969 from sosiouxme/20170801-tolerate-ovs-beyond-36OpenShift Bot2017-08-022-90/+80
|\ \ \ | |/ / |/| | Merged by openshift-bot
| * | package_version check: tolerate release version 3.7Luke Meyer2017-08-022-90/+80
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Addresses issue https://github.com/openshift/openshift-ansible/issues/4967 For now, any version >= 3.6 is handled as if it were 3.6. We may want to keep that or fine-tune it later. Also, the ovs_version check is not updated. This is a post-install health check (does not block install/upgrade) with an update already in progress so will be addressed there.
* | | Merge pull request #4968 from sdodson/cfme-imagesOpenShift Bot2017-08-022-1/+13
|\ \ \ | | | | | | | | Merged by openshift-bot
| * | | Use enterprise images for CFME enterprise deploymentsScott Dodson2017-08-012-1/+13
| | | |
* | | | Missing spaceKarthik Prabhakar2017-08-021-1/+1
| | | |
* | | | Merge pull request #4958 from sdodson/logging-metrics-imagetagScott Dodson2017-08-022-2/+2
|\ \ \ \ | | | | | | | | | | Switch logging and metrics OCP image tag from 3.6.0 to v3.6
| * | | | Switch logging and metrics OCP image tag from 3.6.0 to v3.6Scott Dodson2017-08-012-2/+2
| |/ / /
* | | | Merge pull request #4592 from juanvallejo/jvallejo/verify-fluentd-logging-configOpenShift Bot2017-08-0211-109/+563
|\ \ \ \ | | | | | | | | | | Merged by openshift-bot
| * | | | add fluentd logging driver config checkjuanvallejo2017-08-0111-109/+563
| | |/ / | |/| |
* | | | Merge pull request #4916 from richm/mux-client-modeOpenShift Bot2017-08-028-13/+27
|\ \ \ \ | | | | | | | | | | Merged by openshift-bot
| * | | | use mux_client_mode instead of use_mux_clientRich Megginson2017-08-018-13/+27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Instead of the `openshift_logging_use_mux_client` boolean parameter, use `openshift_logging_mux_client_mode` which will allow us to support different mux client use cases: The value `maximal` will cause Fluentd to perform as much of the processing as possible at the local node. This currently means all of the processing *except* for the Kubernetes metadata processing, which will be done by mux. This is the currently recommended mode to use due to current scaling issues. The value `minimal` means that Fluentd will do *no* processing at all, and send the raw logs to mux for processing. This is currently not recommended to use due to current scaling issues. Ansible will warn you if you try to use this mode. `MUX_ALLOW_EXTERNAL` is no longer needed in the mux dc. mux now always operates to process external logs. The ansible setting `openshift_logging_mux_allow_external` is still required in order to set up the mux service to accept connections from outside of the cluster.
* | | | | Merge pull request #4949 from carlpett/patch-1OpenShift Bot2017-08-011-1/+1
|\ \ \ \ \ | | | | | | | | | | | | Merged by openshift-bot
| * | | | | Fix aws_secret_key checkCalle Pettersson2017-07-311-1/+1
| | | | | |
* | | | | | Merge pull request #4961 from jarrpa/minor-namespace-fixScott Dodson2017-08-011-1/+1
|\ \ \ \ \ \ | | | | | | | | | | | | | | GlusterFS: Check for namespace if deploying a StorageClass
| * | | | | | GlusterFS: Check for namespace if deploying a StorageClassJose A. Rivera2017-08-011-1/+1
| | |_|_|/ / | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1476197 Signed-off-by: Jose A. Rivera <jarrpa@redhat.com>
* | | | | | Merge pull request #4894 from tbielawa/reg-dns-scaleup-bz1469336OpenShift Bot2017-08-015-6/+16
|\ \ \ \ \ \ | |_|_|_|/ / |/| | | | | Merged by openshift-bot
| * | | | | Use existing OPENSHIFT_DEFAULT_REGISTRY setting during masters scaleupTim Bielawa2017-07-285-6/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fixes a bug which reported that AFTER a cluster upgrade from OCP 3.5 to 3.6, any masters which are later added via the scaleup playbooks are setting a value for OPENSHIFT_DEFAULT_REGISTRY which is inconsistent with the already configured masters. * OPENSHIFT_DEFAULT_REGISTRY value is saved from oo_first_master * The new scaleup master has a fact set noting it is a scaleup host * The saved OPENSHIFT_DEFAULT_REGISTRY value is used instead of the default for 3.6, which is to use a hard-coded registry value of 'docker-registry.default.svc:5000' Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1469336
* | | | | | Merge pull request #4700 from sdodson/rate-limitOpenShift Bot2017-08-011-4/+4
|\ \ \ \ \ \ | |_|/ / / / |/| | | | | Merged by openshift-bot
| * | | | | Add rate limit configurabilityScott Dodson2017-07-061-4/+4
| | | | | |
* | | | | | Merge pull request #4912 from abutcher/openshift-node-factsOpenShift Bot2017-08-014-28/+50
|\ \ \ \ \ \ | | | | | | | | | | | | | | Merged by openshift-bot
| * | | | | | Move node facts to new openshift_node_facts role.Andrew Butcher2017-07-274-28/+50
| | | | | | |
* | | | | | | Merge pull request #4900 from ↵OpenShift Bot2017-07-3116-151/+170
|\ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ingvagabund/move-common-openshift-node-systemd-unit-tasks-into-separete-files Merged by openshift-bot
| * | | | | | | move common tasks to a single file included by both systemd_units.ymlJan Chaloupka2017-07-2816-151/+170
| | | | | | | |