Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Default groups.oo_etcd_to_config when setting embedded_etcd in control plane ↵ | Andrew Butcher | 2016-11-10 | 1 | -1/+1 |
| | | | | upgrade. | ||||
* | Merge pull request #2730 from dgoodwin/fix-ha-etcd-backup | Scott Dodson | 2016-11-09 | 1 | -1/+14 |
|\ | | | | | Fix HA etcd upgrade when facts cache has been deleted. | ||||
| * | Fix etcd backup failure due to corrupted facts. | Devan Goodwin | 2016-11-09 | 1 | -1/+14 |
| | | | | | | | | | | | | | | | | Simplest way to reproduce this issue is to attempt to upgrade having removed /etc/ansible/facts.d/openshift.fact. Actual cause in the field is not entirely known but critically it is possible for embedded_etcd to default to true, causing the etcd fact lookup to check the wrong file and fail silently, resulting in no etcd_data_dir fact being set. | ||||
* | | Revert "Revert openshift.node.nodename changes" | Scott Dodson | 2016-11-08 | 2 | -8/+10 |
| | | |||||
* | | Merge pull request #2751 from sdodson/BZ1388253 | Scott Dodson | 2016-11-08 | 2 | -10/+8 |
|\ \ | | | | | | | Revert openshift.node.nodename changes | ||||
| * | | Revert "Fix OpenStack cloud provider" | Scott Dodson | 2016-11-07 | 2 | -10/+8 |
| |/ | | | | | | | This reverts commit 1f2276fff1e41c1d9440ee8b589042ee249b95d7. | ||||
* / | Add missing symlink for node openvswitch oom fix. | Devan Goodwin | 2016-11-08 | 1 | -0/+1 |
|/ | |||||
* | Merge pull request #2670 from abutcher/hosted-registry-insecure | Scott Dodson | 2016-10-26 | 1 | -1/+1 |
|\ | | | | | Bug 1388016 - The insecure-registry address was removed during upgrade | ||||
| * | Default hosted_registry_insecure true when insecure registry present in ↵ | Andrew Butcher | 2016-10-26 | 1 | -1/+1 |
| | | | | | | | | existing /etc/sysconfig/docker. | ||||
* | | Merge pull request #2637 from dgoodwin/ctrl-plane-restart | Scott Dodson | 2016-10-25 | 1 | -9/+6 |
|\ \ | | | | | | | Fix and reorder control plane service restart. | ||||
| * | | Fix and reorder control plane service restart. | Devan Goodwin | 2016-10-21 | 1 | -9/+6 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | This was missed in the standalone upgrade control plane playbook. However it also looks to be out of order, we should restart before reconciling and upgrading nodes. As such moved the restart directly into the control plane upgrade common code, and placed it before reconciliation. | ||||
* | | | [upgrades] Fix containerized node | Scott Dodson | 2016-10-25 | 1 | -2/+0 |
| |/ |/| | | | | | This file was removed and no longer used | ||||
* | | Move infrastructure templates into openshift_hosted_templates role. | Andrew Butcher | 2016-10-21 | 2 | -0/+6 |
| | | |||||
* | | trouble creating service signer while running upgrade dockerized | Henning Fjellheim | 2016-10-21 | 1 | -4/+4 |
| | | |||||
* | | Merge pull request #2623 from rhcarvalho/typos | Andrew Butcher | 2016-10-20 | 1 | -1/+1 |
|\ \ | | | | | | | Fix typos | ||||
| * | | Fix typos | Rodolfo Carvalho | 2016-10-19 | 1 | -1/+1 |
| | | | |||||
* | | | Merge pull request #2621 from dgoodwin/symlink-fix | Scott Dodson | 2016-10-19 | 4 | -14/+10 |
|\ \ \ | |/ / |/| | | Switch from "oadm" to "oc adm" and fix bug in binary sync. | ||||
| * | | Switch from "oadm" to "oc adm" and fix bug in binary sync. | Devan Goodwin | 2016-10-19 | 4 | -14/+10 |
| |/ | | | | | | | | | | | | | | | Found bug syncing binaries to containerized hosts where if a symlink was pre-existing, but pointing to the wrong destination, it would not be corrected. Switched to using oc adm instead of oadm. | ||||
* | | [logging] Fix NFS volume binding | Scott Dodson | 2016-10-17 | 1 | -6/+6 |
| | | |||||
* | | Merge pull request #2608 from abutcher/template-with-items | Scott Dodson | 2016-10-17 | 3 | -3/+3 |
|\ \ | |/ |/| | Template with_items for upstream ansible-2.2 compat. | ||||
| * | Template with_items for upstream ansible-2.2 compat. | Andrew Butcher | 2016-10-14 | 3 | -3/+3 |
| | | |||||
* | | Merge pull request #2605 from sdodson/logging | Scott Dodson | 2016-10-14 | 1 | -6/+6 |
|\ \ | |/ |/| | [logging] Use inventory variables rather than facts | ||||
| * | Use inventory variables rather than facts | Scott Dodson | 2016-10-14 | 1 | -6/+6 |
| | | |||||
* | | Resume restarting node after upgrading node rpms. | Devan Goodwin | 2016-10-14 | 1 | -0/+4 |
| | | | | | | | | | | | | | | Error in commit 245fef16573757b6e691c448075d8564f5d569f4. As it turns out this is the only place a rpm based node can be restarted in upgrade. Restoring the restart but making it conditional to avoid the two issues reported with out of sync node restarts. | ||||
* | | upgrade: Don't check avail docker version if not already installed. | Devan Goodwin | 2016-10-14 | 1 | -2/+4 |
|/ | |||||
* | Stop restarting node after upgrading master rpms. | Devan Goodwin | 2016-10-12 | 1 | -3/+0 |
| | | | | | | | | | | This looks to be causing a customer issue where some HA upgrades fail, due to a missing EgressNetworkPolicy API. We update master rpms, we don't restart services yet, but then restart node service which tries to talk to an API that does not yet exist. (pending restart) Restarting node here is very out of place and appears to not be required. | ||||
* | Merge pull request #2441 from dgoodwin/34-upgrade-improvements | Andrew Butcher | 2016-10-04 | 17 | -408/+447 |
|\ | | | | | 3.4 Upgrade Improvements | ||||
| * | Fix bug with service signer cert on upgrade. | Devan Goodwin | 2016-09-29 | 2 | -1/+7 |
| | | | | | | | | | | | | It is invalid Ansible to use a when on an include that contains plays, as it cannot be applied to plays. Issue filed upstream for a better error, or to get it working. | ||||
| * | Use pre_upgrade tag instread of a dry run variable. | Devan Goodwin | 2016-09-29 | 1 | -1/+1 |
| | | |||||
| * | Move etcd backup from pre-upgrade to upgrade itself. | Devan Goodwin | 2016-09-29 | 2 | -87/+86 |
| | | |||||
| * | Allow a couple retries when unscheduling/rescheduling nodes in upgrade. | Devan Goodwin | 2016-09-29 | 1 | -0/+12 |
| | | | | | | | | | | | | | | | | This can fail with a transient "object has been modified" error asking you to re-try your changes on the latest version of the object. Allow up to three retries to see if we can get the change to take effect. | ||||
| * | Skip the docker role in early upgrade stages. | Devan Goodwin | 2016-09-29 | 2 | -5/+6 |
| | | | | | | | | | | | | | | | | This improves the situation further and prevents configuration changes from accidentally triggering docker restarts, before we've evacuated nodes. Now in two places, we skip the role entirely, instead of previous implementation which only skipped upgrading the installed version. (which did not catch config issues) | ||||
| * | Allow filtering nodes to upgrade by label. | Devan Goodwin | 2016-09-29 | 7 | -13/+55 |
| | | |||||
| * | Allow customizing node upgrade serial value. | Devan Goodwin | 2016-09-29 | 1 | -1/+3 |
| | | |||||
| * | Split upgrade for control plane/nodes. | Devan Goodwin | 2016-09-29 | 10 | -103/+89 |
| | | |||||
| * | Verify masters are upgraded before proceeding with node only upgrade. | Devan Goodwin | 2016-09-28 | 1 | -3/+0 |
| | | |||||
| * | Attempt to tease apart pre upgrade for masters/nodes. | Devan Goodwin | 2016-09-28 | 8 | -305/+251 |
| | | |||||
| * | Split upgrade entry points into control plane/node. | Devan Goodwin | 2016-09-28 | 2 | -1/+49 |
| | | |||||
| * | Reunite upgrade reconciliation gating with the play it gates on. | Devan Goodwin | 2016-09-28 | 1 | -17/+18 |
| | | |||||
| * | Drop atomic-enterprise as a valid deployment type in upgrade. | Devan Goodwin | 2016-09-28 | 1 | -2/+2 |
| | | |||||
| * | Stop guarding against pacemaker in upgrade, no longer necessary. | Devan Goodwin | 2016-09-28 | 1 | -8/+0 |
| | | |||||
| * | Support openshift_upgrade_dry_run=true for pre-upgrade checks only. | Devan Goodwin | 2016-09-28 | 1 | -0/+7 |
| | | |||||
* | | Merge pull request #2475 from smunilla/registry_all_the_time | Scott Dodson | 2016-09-30 | 1 | -102/+1 |
|\ \ | | | | | | | Install Registry by Default | ||||
| * | | Further secure registry improvements | Andrew Butcher | 2016-09-29 | 1 | -128/+0 |
| | | | | | | | | | | | | | | | | | | - Default to hosted_registry_insecure=False - Add openshift ca to system ca-trust. - Update ca trust in openshift_node_certificates rather than docker_ca_trust | ||||
| * | | Delgate handlers to first master | Samuel Munilla | 2016-09-29 | 1 | -0/+4 |
| | | | |||||
| * | | Secure registry improvements. | Andrew Butcher | 2016-09-29 | 2 | -102/+125 |
| | | | | | | | | | | | | | | | | | | * Convert oc template calls to jsonpath. * Wait for deployments to finish before restarting docker. * Re-organize node ca configuration. | ||||
| * | | Install Registry by Default | Samuel Munilla | 2016-09-29 | 1 | -9/+9 |
| | | | | | | | | | | | | | | | | | | | | | Instead of restricting cockpit-ui to Atomic Registry installations, install it by default everywhere. Fixes Bug 1371459 | ||||
* | | | Merge pull request #2532 from abutcher/logging-deploy-bool | Scott Dodson | 2016-09-29 | 1 | -2/+2 |
|\ \ \ | |/ / |/| | | Default openshift_hosted_{logging,metrics}_deploy to false. | ||||
| * | | Default openshift_hosted_{logging,metrics}_deploy to false. | Andrew Butcher | 2016-09-29 | 1 | -2/+2 |
| | | | |||||
* | | | Merge pull request #2529 from abutcher/play-names | Scott Dodson | 2016-09-29 | 2 | -2/+3 |
|\ \ \ | |/ / |/| | | Update play names for consistency. |