summaryrefslogtreecommitdiffstats
path: root/playbooks
Commit message (Collapse)AuthorAgeFilesLines
* Minor upgrade improvements.Devan Goodwin2015-11-062-7/+12
| | | | | | | Skip some 3.1 checks if doing a 3.0.x to 3.0.2 upgrade. Improve error message when oc whoami fails (i.e. openshift is down) during pre-upgrade checks, rather than assuming the binary doesn't exist.
* remove debug statementJason DeTiberus2015-11-061-2/+0
|
* Fix removal of kubernetesMasterConfig.apiLevelsJason DeTiberus2015-11-061-1/+1
|
* Fix issues related to upgrade packages being unavailableJason DeTiberus2015-11-062-6/+16
|
* Merge pull request #51 from dgoodwin/pre-upgradeJason DeTiberus2015-11-061-1/+4
|\ | | | | Pre upgrade
| * Fix creation of origin symlink when dir already exists.Devan Goodwin2015-11-061-1/+4
| |
* | Merge remote-tracking branch 'origin/upgradeFix2' into upgradeFix2Jason DeTiberus2015-11-062-3/+243
|\ \
| * | Merge pull request #50 from dgoodwin/pre-upgradeJason DeTiberus2015-11-062-0/+192
| |\| | | | | | | Run pre-upgrade script on nodes
| | * Add pre-upgrade script to be run on first master.Devan Goodwin2015-11-062-0/+192
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Script currently just checks for port names that are no longer valid. In theory other checks may be added to this script in the future. Script was originally written by Steve Milner and Andy Goldstein. If the script fails, ansible seems to handle this nicely by default, exiting the upgrade and displaying stderr and stdout, both of which contain useful info on what the problem was.
| * | Start to handle pacemaker ha during upgradeAndrew Butcher2015-11-061-3/+51
| |/
* / Fix apiLevels modificationsJason DeTiberus2015-11-061-7/+3
|/
* add master_hostnames definition for upgradeJason DeTiberus2015-11-051-0/+4
|
* Additional upgrade enhancementsJason DeTiberus2015-11-053-72/+101
| | | | | | | | - rework the version checking - provide better safety if the apiLevel attributes are missing - ensure a list of api levels are present - remove a list of api levels - pylint fixes
* Handle backups for separate etcd hosts if necessary.Devan Goodwin2015-11-051-9/+27
|
* Further upgrade improvementsJason DeTiberus2015-11-051-46/+168
| | | | | - Restart masters post reconcile - generate missing master certs and sync to masters
* Upgrade improvementsDevan Goodwin2015-11-052-75/+56
| | | | | | | | | | - Push config dir logic out of module and use host variables instead. - Backup master config with ansible utility. - Add error handling for the upgrade config module. - Add verbose option to installer. - Return details on what we changed when upgrading config. - Cleanup use of first master. - Don't install upgrade rpms to check what version we'll upgrade to.
* Merge pull request #608 from abutcher/native-haBrenton Leanhardt2015-11-055-5/+96
|\ | | | | Native Support for Multi-Master HA
| * Various HA changes for pacemaker and native methods.Andrew Butcher2015-11-054-15/+59
| |
| * default to source persistence for haproxyJason DeTiberus2015-11-041-1/+1
| |
| * more tweaksJason DeTiberus2015-11-041-0/+1
| |
| * more tweaksJason DeTiberus2015-11-041-35/+9
| |
| * additional ha related updatesJason DeTiberus2015-11-043-6/+52
| |
| * additional native ha changesAndrew Butcher2015-11-041-5/+5
| |
| * Start of true master haJason DeTiberus2015-11-041-0/+26
| |
* | The uninstall playbook needs to remove /run/openshift-sdnBrenton Leanhardt2015-11-051-2/+6
| | | | | | | | | | If this isn't removed lbr0 won't be recreated by the Node after the following install.
* | Remove pacemaker bits.Andrew Butcher2015-11-041-0/+6
|/
* Override hosts deployment_type fact for version we're upgrading to.Devan Goodwin2015-11-041-0/+10
|
* Pylint fixes for config upgrade module.Devan Goodwin2015-11-041-10/+13
|
* Disable proxy cert config upgrade until certs being generated.Devan Goodwin2015-11-041-5/+6
|
* Merge remote-tracking branch 'upstream/master' into upgradeDevan Goodwin2015-11-0411-75/+98
|\
| * Merge pull request #761 from smunilla/scaleupBrenton Leanhardt2015-11-049-75/+83
| |\ | | | | | | openshift-cluster: Scaleup
| | * Refactor common group evaluation to avoid duplicationJason DeTiberus2015-11-049-139/+77
| | |
| | * common/openshift-cluster: Scaleup playbookSamuel Munilla2015-11-041-0/+70
| | | | | | | | | | | | New playbook to add nodes to a cluster without altering the masters
| * | Merge pull request #767 from abutcher/custom-certsBrenton Leanhardt2015-11-041-0/+9
| |\ \ | | |/ | |/| Add custom certificates to serving info in master configuration.
| | * Filter internal hostnames from the list of parsed names.Andrew Butcher2015-11-031-1/+1
| | |
| | * Add custom certificates to serving info in master configuration.Andrew Butcher2015-11-031-0/+9
| | |
| * | Disable requiretty for only the openshift usererror102015-11-011-1/+6
| | | | | | | | | Use write_files to disable requiretty for the openshift user as suggested by @detiberm, fixes #773
| * | Don't require tty to run sudoerror102015-10-301-0/+1
| |/ | | | | Set Defaults !requiretty so that ansible can run sudo without a terminal. Fixes #773
* | Fix bug with not upgrading openshift-master to atomic-openshift-master.Devan Goodwin2015-11-041-14/+5
| | | | | | | | | | | | | | | | Removing the full call to config resulted in rpms not getting upgraded. Config was doing a yum update of everything, which picks up the atomic-openshift-master obsoleting openshift-master. The actual yum call changed here would not. Instead we switch to a direct call to yum which correctly picks up the obsoletes and updates to atomic-openshift packages.
* | Merge branch 'config-upgrade' into upgradeDevan Goodwin2015-11-043-11/+128
|\ \
| * | Move config upgrade to correct place, fix node facts.Devan Goodwin2015-11-032-12/+23
| | |
| * | Add in proxyClientInfo if missing during config upgrade.Devan Goodwin2015-11-031-3/+9
| | |
| * | Implement master-config.yaml upgrade for v1beta3 apiLevel removal.Devan Goodwin2015-11-032-12/+66
| | |
| * | Merge branch 'upgrade' into config-upgradeDevan Goodwin2015-11-031-0/+43
| |\ \
| * | | initial module frameworkJason DeTiberus2015-11-021-0/+62
| | |/ | |/|
* | | Use the base package upgrade version so we can check things earlier.Devan Goodwin2015-11-041-23/+41
| | |
* | | Skip fail if enterprise deployment type depending on version.Devan Goodwin2015-11-041-0/+8
| | |
* | | Add debug output for location of etcd backup.Devan Goodwin2015-11-031-0/+3
| |/ |/|
* | Merge remote-tracking branch 'upstream/master' into upgradeDevan Goodwin2015-11-024-6/+55
|\|
| * Attempt to remove the various interfaces left over from an installBrenton Leanhardt2015-10-301-0/+14
| | | | | | | | | | | | | | | | Removing br0 won't work if openvswitch has already been stopped. ovs-system will be left around but that is just a problem with openvswitch's packaging. That device is apparently not even used. In the worst case a manual reboot will remove all of these devices.