* vendoring kubernetes/client-go
* removing code that was added to make sure vendoring worked
* copying code from kubernetes/kubernetes to do drain command
Still need to make it testable. and then test it.
* making the code unit testable and writing unit tests.
* needed to merge manually for rebase
* updating upgrade command to validate location parameter and updating e2e test file to pass in location
* fixing style mistake
* More style fixes
* fixing some build breaks
had been using make test to build and that doesn't build the commands
* fixing upgrade api to accept kube config
Also commented out drain due to issues found. Opened issues for them
#1548#1547
* changing string constant in test to show is a kubeconfig being passed in
* adding missing import
* reverting a strange change that snuck in.
* not introducing a new error path in tests
* making timeout honored and passed as a parameter so it will be configurable in the RP
* deprecate orchestratorRelease, and make build pass
* fix tests after deprecating orchestratorRelease
* fix documents
* fix e2e tests
* remove 20170831 kubernetesRelease
* let vlabs take orchestratorRelease
* vlabs still use orchestratorRelease for test api model
* fix lint
* changed ubuntu image to 16.04.201706191
* Replaced ubuntu image with 16.04.201708151 in more files
* Updated Ubuntu image for German cloud
* Updated Ubuntu image in example template
* Updated Ubuntu image in kubernbetes update example
* Set etcd restart to always
* Reverted change for etcd restart
Change done in another branch
* Revert German version back to 201701130
* Remove calicoConfigURL
* upgrade to 2.6.0 calico
* update to 2.6.1 calico
* use calico15 addon w K8s 1.6
* remove CalicoConfigDownloadURL
* calico version switch per K8s version
* The linux definition now builds a more feature complete kubernetes cluster. This means we can provision fewer clusters while testing more surface area
* Added support for RHEL
* Add new distro element to API
* Define and set RHEL image
* Modify templates to support RHEL
* Run gofmt
* Convert Distro field
* Example for Docker swarm mode on RHEL
* Point config script URL to acs-engine repo
* Added script root URL parameter
* Move distro parameter to master/agent config
* Restore OS image parameters
* eliminated duplicated structures
* added run-time check for upgradable clusters
* updated k8sUpgrade.json
* replaced proprietary implementation of orch. versions with semver package
* updated API loader for the upgrade operation in ContainerService
* added pre-validation for the upgrade
* removed experimental flag from the upgrade operation
* deleted redundant UpgradeContainerService. Use OrchestratorProfile instead
* adding support to add query parameters to calls to get resource urls
this allows the use of sas keys in calls to azure storage. Fixed a
couple of other issues I ran into while using this.
* update to get post provision scripts working for DCOS and Swarm
* white space and explanation comment
* ensure k8s custom script provision runs
I was having issues with waagent restarting in the middle of the custom
script extension running.
* some clean up on examples
* rebasing had to handle a couple merge conflicts
* Added extensions support
* updated supported-orchestrators to include dcos 1.9
* Add OMS Agent for Kubernetes
* Updated the extensions type and typeHandlerVersion to match the version now used in the templates. Added an example for oms.
* Add VAMP DC/OS extension
* updating documentation for changes I made to example extensions
* updating dcos deployments to happen after all dcos masters have finished and consensus achieved
* allowing agents or masters to get extensions
* adding single or all support
* handling vmss, fixing a bug, and allowing passing index in extension parameters
* changing extension names to match what is in the template so that they can run
* addressing PR feedback
* updating documentation to state this doesn't work for k8s
* adding preprovion exetnsions to api model
* getting preprovision scripts working for swarm
* get pre provision scripts working in all orchestrators and with indexes
* adding sample preprovision script, updating documentation, simplifying default root url logic
* style fix
* fixing broken unit test
* adding unit test for prerprovision extension
* addressing PR feedback and adding examples for future E2E tests
* removing test files
* fixing example mistakes
* Add maxPods to Kubernetes configuration
* Group Kubernetes config examples and add maxPods example
* Update E2E configs to point to the new example files
* introduce OrchestratorVersionHint
* replace places with OrchestratorVersion with OrchestratorVersionHint, and build pass
* fill in OrchestratorVersion when converttoapi
* fix test
* use go-version instead of VersionOrdinal
* use k8sVersion instead of k8sVersionHint in comparison
* remove duplicate consts, use common
* split KubeImagee into images and configs
* fix more KubernetesVersion referring
* style and deadcode fixes
* kube config validate needs version hint string
* need string literal to refer to v “1.5.3”
* enforcing default “” orchestratorVersion in apiloader
* gomega unused in engine_test.go
* s/VersionHint/Release
replacing the term “VersionHint” with “Release” throughout
* chore(gofmt)
* chore(glide) go-version
* rationalized KubeConfigs
- added pkg/acsengine/api.KubernetesReleaseToVersion as a map from major.minor “release” to specific “version”; replaced old usages of `api.KubeImages[<version>][”version”] with `api.KubernetesReleaseToVersion[<version>]`
- removed api.KubeImages
- all version-specific k8s config now derive from pkg/acsengine.KubeConfigs
* calico 1.5 logic now uses OrchestratorRelease
* make upgrade scenario working again
* fix upgrade test
* fix more for upgrade scenario
* fix release to versin mapping to 1.7.2
* adding test files for k8s upgrade
* set azure authentication
* set EXPECTED_ORCHESTRATOR_VERSION
* allow delay after upgrade to let cluster settle up
* added scenarios for windows and hybrid clusters
* added README
* added test deployment configuration
* Added minimal rbac setup for k8s.
* Make heapster rolebinding.
* Add kube-dns rolebindings.
* Smaller fixes,
* Conditionalize enabling of RBAC
Add rbac cluster definition example
Update kube addon manager to support clusterrolebinding creation
Update kube-addon-manager to use service account
Update API server with RBAC parameter
Update Controller Manager to use separate accounts per controller
Update heapster to support RBAC
Update dashboard to support RBAC
Add enableRbac field to ClusterConfig and add docs
* remove service account from addon manager
* Add kube-dns rolebindings.
* Smaller fixes,
* Conditionalize enabling of RBAC
Add rbac cluster definition example
Update kube addon manager to support clusterrolebinding creation
Update kube-addon-manager to use service account
Update API server with RBAC parameter
Update Controller Manager to use separate accounts per controller
Update heapster to support RBAC
Update dashboard to support RBAC
Add enableRbac field to ClusterConfig and add docs
* remove service account from addon manager
* update heapster deployment
* Fix some merge conflicts.
* Added test for certificate creation with and without organisation.
* update CR and CRB to support heapster nanny rather than use the bootstrapped CR
* removed unrelated modifications to addons
* re-add resource limits and remove quotes
* Make heapster rolebinding.
* Add kube-dns rolebindings.
* Smaller fixes,
* Conditionalize enabling of RBAC
Add rbac cluster definition example
Update kube addon manager to support clusterrolebinding creation
Update kube-addon-manager to use service account
Update API server with RBAC parameter
Update Controller Manager to use separate accounts per controller
Update heapster to support RBAC
Update dashboard to support RBAC
Add enableRbac field to ClusterConfig and add docs
* remove service account from addon manager
* Make heapster rolebinding.
* Smaller fixes,
* Conditionalize enabling of RBAC
Add rbac cluster definition example
Update kube addon manager to support clusterrolebinding creation
Update kube-addon-manager to use service account
Update API server with RBAC parameter
Update Controller Manager to use separate accounts per controller
Update heapster to support RBAC
Update dashboard to support RBAC
Add enableRbac field to ClusterConfig and add docs
* remove service account from addon manager
* update heapster deployment
* Fix some merge conflicts.
* update CR and CRB to support heapster nanny rather than use the bootstrapped CR
* removed unrelated modifications to addons
* re-add resource limits and remove quotes
* rebase master heapster updates, add 1.5 rbac apis
* manifest updates
* update addon-manager version across all releases, update kube-dns 1.5
* remove templates.go
* fixes based on review
* add docs
* post rebase updates
* misc addon fixes post rebase
* add newline to example and fix docs with master
* yaml formatting fix
* Add ServicePrincipalProfile.KeyvaultSecretRef for KeyVault secret reference
- with this change, ServicePrincipalProfile.secret will be added to template parameters as is without being tranformed into Keyvault secret reference.
- to use keyvault secret reference, use ServicePrincipalProfile.KeyvaultSecretRef instead
* refactored to address comments
* fixing the tests
* converted —node-status-update-frequency
nodeStatusUpdateFrequency is now delivered via KubernetesConfig
* all recent k8s config params are api-model-enabled
* rationalizing config outputs as string params
* added k8s version-specific validation
added regression tests per test category
added error code for FaultDomainCount into reporting
added test scenario for k8s/dcos/swarmmode master with storage account
added deployment test specs for k8s windows
added configurable resource group name prefix for automatic tests
* acs-engine configs for backoff
* errata
* large cluster support in 1.6.6 only at this point
* add custom data vars to master as well
* updated templates.go
* kube-controller-manager var substitution is unique
* updated generated template
* moving large cluster to examples/largeclusters
* working pool names for large cluster example
dmitsh
Jack
Jingtao Ren
Cecile Robert-Michon
David Tesar
Jack Francis
Weinong Wang
Thomas Conté
Jonathan Chauncey
itowlson
Julien Stroheker
Nicholas Althoff
Onur Filiz
Steve Kriss
Anubhuti Manohar
Jason Hansen
Bruce Campbell
David Porter
Lachlan Evenson
uday31in
Cole Mickens