summaryrefslogtreecommitdiffstats
path: root/roles
Commit message (Collapse)AuthorAgeFilesLines
* Revert "GlusterFS: Remove image option from heketi command"Michael Hanselmann2018-01-301-1/+1
| | | | | | | | | | | | | | | | | | | | | This reverts commit 3d3853836d71c37a9c81aea8e606c94052439789 which was supposedly a fix for an issue with topology initialization, but that turned out to be a separate issue fixed in commit b8879b5c6731bd51e590b470b1a51fce73db0ee1. By default "heketi-cli setup-openshift-heketi-storage" uses the "heketi/heketi:dev" image while, for OpenShift Container Platform installations, the default image for all other operations is "rhgs3/rhgs-volmanager-rhel7". In environments where only a limited set of Docker registries is available, i.e. only the Red Hat registry but not Docker Hub, using "heketi/heketi:dev" does not work or would require manual intervention. The discussion surrounding the removal of "--image" in PR#5769 involved statements that "--image" was removed from Heketi. As of January 30, 2018 the master branch contains the option, as does the upstream Docker image (Heketi version v5.0.0-225-gffb9fea) and the aforementioned "rhgs3/rhgs-volmanager-rhel7" image. Tested with OpenShift Container Platform 3.7.23.
* Merge pull request #6923 from openshift/revert-6913-revert-6830-registry_addrOpenShift Merge Robot2018-01-291-1/+1
|\ | | | | | | | | | | | | | | | | Automatic merge from submit-queue. Revert "Revert "use non-deprecated REGISTRY_OPENSHIFT_SERVER_ADDR variable to set the registry hostname"" Reverts openshift/openshift-ansible#6913 image-registry has been fixed to properly respect this variable now. re-enabling this change.
| * Revert "Revert "use non-deprecated REGISTRY_OPENSHIFT_SERVER_ADDR variable ↵Ben Parees2018-01-291-1/+1
| | | | | | | | to set the registry hostname""
* | Merge pull request #6860 from jpeeler/aggregate-name-changeScott Dodson2018-01-294-106/+97
|\ \ | | | | | | Update catalog role patching to use new names
| * | Change catalog roles install to use aggregationJeff Peeler2018-01-244-106/+97
| | | | | | | | | | | | | | | | | | | | | | | | Recently it was changed for openshift cluster roles to be reconciled via the aggregate-to* label, so this is now updated. Rather than generating portions of the file, simply use the auth reconciler and let it handle any necessary role updates. Also, since it is theoretically possible for conflicts to occur, retry reconciliation several times.
* | | Merge pull request #6889 from mazzystr/fix-port-alignmentOpenShift Merge Robot2018-01-294-63/+76
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Automatic merge from submit-queue. Fix misaligned ports for sg,elb,api Fix to issue [6867](https://github.com/openshift/openshift-ansible/issues/6867) The scope of this patch widened slightly to consolidate some AWS ELB vars/dicts into the single dictionary due to Health check being sourced by all ELBs. By default health check needs to ping 8443 for master ELBS and 443 for infra ELBs. Consolidating allows for changes to the behavior of one ELB without affecting the others. @sdodson @kwoodson
| * | | Fix misaligned ports for sg,elb,apiChris Callegari2018-01-254-63/+76
| | | |
* | | | Merge pull request #6215 from sdodson/bz1511870OpenShift Merge Robot2018-01-291-0/+4
|\ \ \ \ | |_|_|/ |/| | | | | | | | | | | | | | | | | | | | | | | | | | | Automatic merge from submit-queue. Add kube service ipaddress to no_proxy list Computed as the first ip address in the portal_net Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1511870
| * | | Make IP object a stringFabian von Feilitzsch2018-01-241-1/+2
| | | |
| * | | Add kube service ipaddress to no_proxy listScott Dodson2018-01-241-0/+3
| | | | | | | | | | | | | | | | Computed as the first ip address in the portal_net
* | | | Merge pull request #6909 from spadgett/detect-croOpenShift Merge Robot2018-01-291-1/+6
|\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Automatic merge from submit-queue. Bug 1539182: Detect if ClusterResourceOverrides enabled during console install Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1539182 /assign @sdodson /cc @jwforres @jupierce
| * | | | Bug 1539182: Detect if ClusterResourceOverrides enabled during console installSamuel Padgett2018-01-271-1/+6
| | | | | | | | | | | | | | | | | | | | Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1539182
* | | | | Merge pull request #6811 from mjudeikis/prometheus-new-templateOpenShift Merge Robot2018-01-297-142/+273
|\ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Automatic merge from submit-queue. Prometheus new template rebase Updating Prometheus for new templates/example. 1. New scraping rules, fixes 2. exposed alert manager 3. clean ansible 4. add a custom annotation for routes (in the example when AVI router in use we need to be able to add custom annotations) 5. Externalise some of the configs Still work in progress... FIY: @zgalor
| * | | | | Rebase Prometheus example for new scrape endpoints and expose alert managerMangirdas2018-01-287-142/+273
| | | | | |
* | | | | | Revert "use non-deprecated REGISTRY_OPENSHIFT_SERVER_ADDR variable to set ↵Ben Parees2018-01-271-1/+1
| |/ / / / |/| | | | | | | | | | | | | | the registry hostname"
* | | | | Merge pull request #6905 from mgugino-upstream-stage/fix-crt-typoOpenShift Merge Robot2018-01-271-1/+1
|\ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Automatic merge from submit-queue. Fix container_runtime variable typo 'Centos' should be 'CentOS' as it corresponds to the ansible variable that determines the linux distro.
| * | | | | Fix container_runtime variable typoMichael Gugino2018-01-261-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 'Centos' should be 'CentOS' as it corresponds to the ansible variable that determines the linux distro.
* | | | | | Merge pull request #6896 from mgugino-upstream-stage/fix-upgrade-image-tagOpenShift Merge Robot2018-01-262-2/+5
|\ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Automatic merge from submit-queue. Correct 3.7 to 3.9 upgrade openshift_image_tag 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
| * | | | | | Correct 3.7 to 3.9 upgrade openshift_image_tagMichael Gugino2018-01-262-2/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 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
* | | | | | | Merge pull request #6840 from yocum137/rm_origin-componentsOpenShift Merge Robot2018-01-2612-7/+438
|\ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Automatic merge from submit-queue. moving files to their correct <role>/files location The openshift_web_console and template_service_broker roles are putting 'files' outside the openshift-ansible roles they belong to.
| * \ \ \ \ \ \ Merge branch 'master' of github.com:openshift/openshift-ansible into ↵Dan Yocum2018-01-2655-325/+1185
| |\ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | rm_origin-components
| * | | | | | | | files in ansible roles do not need to have the path specified to themDan Yocum2018-01-245-7/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | when referenced by a builtin module, i.e., copy:
| * | | | | | | | Merge branch 'master' into rm_origin-componentsDan Yocum2018-01-2470-1047/+5400
| |\ \ \ \ \ \ \ \ | | | |_|_|_|_|_|/ | | |/| | | | | |
| * | | | | | | | moving files to their correct <role>/files dir for the openshift_web_console ↵Dan Yocum2018-01-239-2/+430
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | and template_service_broker roles
* | | | | | | | | Merge pull request #6888 from kwoodson/fix_overlay_nodriveOpenShift Merge Robot2018-01-262-1/+5
|\ \ \ \ \ \ \ \ \ | |_|_|/ / / / / / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Automatic merge from submit-queue. Removing dependency on the extra storage device. This pull request addresses the need to have configure docker storage to overlay without an extra drive.
| * | | | | | | | Removing dependency on the extra stroage device.Kenny Woodson2018-01-252-1/+5
| | | | | | | | |
* | | | | | | | | Merge pull request #6871 from joelddiaz/oc_group_ignore_usersOpenShift Merge Robot2018-01-262-2/+2
|\ \ \ \ \ \ \ \ \ | |_|_|_|_|/ / / / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Automatic merge from submit-queue. ignore 'users' field in oc_group module oc_group doesn't manage the list of users assigned to a group (oc_user does). so when doing an oc_group: state: present name: groupA on a pre-existing group 'groupA' with already defined users, oc_group would detect a difference and blow away the existing users fix this by adding 'users' to the list of fields to ignore when determining whether there are object differences
| * | | | | | | | ignore 'users' field in oc_group moduleJoel Diaz2018-01-252-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | oc_group doesn't manage the list of users assigned to a group (oc_user does). so when doing an oc_group: state: present name: groupA on a pre-existing group 'groupA' with already defined users, oc_group would detect a difference an blow away the existing users fix this by adding 'users' to the list of fields to ignore when determining whether there are object differences
* | | | | | | | | Merge pull request #6862 from ewolinetz/deprecate_use_callbackOpenShift Merge Robot2018-01-262-4/+7
|\ \ \ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Automatic merge from submit-queue. Updating deprecations to use callback plugin Instead of just pausing during the run and relying on users to see it, instead format and print the deprecated variables at the end of play instead like this: ``` Initialization : Complete (0:00:23) [DEPRECATION WARNING]: The following are deprecated variables and will be no longer be used in the next minor release. Please update your inventory accordingly. openshift_hosted_logging_deploy openshift_hosted_metrics_deployer_prefix ``` The plugin changes are the same as the changes made in https://github.com/openshift/openshift-ansible/pull/6734
| * | | | | | | | | Updating deprecations to use callback pluginEric Wolinetz2018-01-253-4/+11
| | | | | | | | | |
* | | | | | | | | | Merge pull request #6882 from spadgett/console-node-selectorOpenShift Merge Robot2018-01-261-1/+1
|\ \ \ \ \ \ \ \ \ \ | |_|_|_|_|_|/ / / / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Automatic merge from submit-queue. Run console pods on the master Update the node selector for the console so that it runs on the masters. /assign @sdodson
| * | | | | | | | | Run console pods on the masterSamuel Padgett2018-01-251-1/+1
| | |_|_|_|_|_|_|/ | |/| | | | | | | | | | | | | | | | | | | | | | | | | Update the node selector for the console so that it runs on the masters.
* | | | | | | | | Merge pull request #6821 from spadgett/remove-asset-configScott Dodson2018-01-264-2/+24
|\ \ \ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | Remove old assetConfig from master-config.yaml
| * | | | | | | | | Remove old assetConfig from master-config.yamlSamuel Padgett2018-01-253-2/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 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
| * | | | | | | | | Don't emit assetConfig on 3.9Scott Dodson2018-01-251-0/+3
| |/ / / / / / / /
* | | | | | | | | Merge pull request #6880 from sosiouxme/20180115-docker-image-locationsOpenShift Merge Robot2018-01-262-13/+47
|\ \ \ \ \ \ \ \ \ | |_|_|_|_|_|_|/ / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Automatic merge from submit-queue. docker_image_availability: containerized image overrides Fix https://bugzilla.redhat.com/show_bug.cgi?id=1538806 and related issues by handling the various containerized image variables.
| * | | | | | | | docker_image_availability: containerized overridesLuke Meyer2018-01-252-13/+47
| |/ / / / / / / | | | | | | | | | | | | | | | | | | | | | | | | Respect container image name override variables.
* | | | | | | | Merge pull request #6857 from mtnbikenc/fix-1537946OpenShift Merge Robot2018-01-251-1/+2
|\ \ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Automatic merge from submit-queue. [1537946] Correct conditional check for GlusterFS IPs Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1537946
| * | | | | | | | [1537946] Correct conditional check for GlusterFS IPsRussell Teague2018-01-241-1/+2
| | | | | | | | |
* | | | | | | | | Merge pull request #6875 from bparees/syncOpenShift Merge Robot2018-01-2521-127/+746
|\ \ \ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Automatic merge from submit-queue. sync imagestreams+templates from origin master for v3.9 fixes https://github.com/openshift/openshift-ansible/issues/6872
| * | | | | | | | | sync imagestreams+templates from origin master for v3.9Ben Parees2018-01-2521-127/+746
| | |/ / / / / / / | |/| | | | | | |
* | | | | | | | | Merge pull request #6859 from abutcher/bz1536217OpenShift Merge Robot2018-01-251-0/+53
|\ \ \ \ \ \ \ \ \ | |_|_|_|_|/ / / / |/| | | | | | | | | | | | | | | | | | | | | | | | | | Automatic merge from submit-queue. Bug 1536217: Need to validate etcd serving certs before 3.9 upgrade
| * | | | | | | | Redeploy etcd certificates during upgrade when etcd hostname not present in ↵Andrew Butcher2018-01-241-0/+53
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | etcd serving cert SAN.
* | | | | | | | | Merge pull request #6878 from mtnbikenc/fix-1502838OpenShift Merge Robot2018-01-251-3/+4
|\ \ \ \ \ \ \ \ \ | |_|/ / / / / / / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Automatic merge from submit-queue. [1502838] Correct certificate alt name parsing Certificates may have alternate names specified, which may contain different name types. Only 'DNS' alternate types should be parsed. X509v3 Subject Alternative Name: email:hostmaster@example.com, DNS:host.example.com Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1502838
| * | | | | | | | [1502838] Correct certificate alt name parsingRussell Teague2018-01-251-3/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Certificates may have alternate names specified, which may contain different name types. Only 'DNS' alternate types should be parsed. X509v3 Subject Alternative Name: email:hostmaster@example.com, DNS:host.example.com Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1502838
* | | | | | | | | Merge pull request #6865 from giuseppe/node-specify-bind-optionScott Dodson2018-01-251-0/+1
|\ \ \ \ \ \ \ \ \ | |/ / / / / / / / |/| | | | | | | | node: specify bind option to /root/.docker
| * | | | | | | | node: specify bind option to /root/.dockerGiuseppe Scrivano2018-01-251-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Without the option, runc fails with "no such device" when trying to create the mount point. Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1534933 Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
* | | | | | | | | Merge pull request #6734 from ewolinetz/logging_red_restart_checkScott Dodson2018-01-253-71/+97
|\ \ \ \ \ \ \ \ \ | |/ / / / / / / / |/| | | | | | | | Only automatically restart if cluster is in yellow or green state
| * | | | | | | | Only automatically restart if cluster is in yellow or green stateEric Wolinetz2018-01-243-71/+97
| |/ / / / / / /
* | | | | | | | Merge pull request #6841 from sdodson/bz1491100OpenShift Merge Robot2018-01-251-0/+2
|\ \ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Automatic merge from submit-queue. Set a default for openshift_hosted_registry_storage_azure_blob_realm Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1491100