* Improved coverage of checkpoint.go to 96.4%
* Improved coverage of checkpoint.go to 96.4%
* Fixed unit tests in checkpoint_test.go and conf_test.go
* Removed unnecessary comments
* improved conf code coverage by an amount that is greater than 0!
* improved coverage, but line 144 of conf.go needs a look
* Added unit tests to multus and types, also fixed a bug in conf.go
* increased code coverage in multus.go and conf.go, also added bug fixes and formatting
* hopefully resolved merge conflicts
* addressed all comments in review
* changed 'thejohn' to '_not_type' for readability
* Added network status annotations section to quickstart and added more unit tests
* added more tests to k8sclient.go
* added another test to k8sclient.go
* Added new function to testing.go and cleaned up tests
* Added a test for GetLoggingLevel
* Added up to 96% coverage for checkpoint.go
* Improved coverage of checkpoint.go to 96.4%
* Improved coverage of checkpoint.go to 96.4%
* Adding changes so i will have them saved for my remote fork thingy
* Fixed unit tests in checkpoint_test.go and conf_test.go
* Removed unnecessary comments
* improved conf code coverage by an amount that is greater than 0!
* improved coverage, but line 144 of conf.go needs a look
* Added unit tests to multus and types, also fixed a bug in conf.go
* added label to import types/020 in types.go
* hopefully resolved merge conflicts
* increased code coverage in multus.go and conf.go, also added bug fixes and formatting
* addressed all comments in review
* Updated testing.go with better comments
* changed 'thejohn' to '_not_type' for readability
* added additional unit tests
* added tests to kubeletclient.go
* added more unit tests to k8sclient.go and kubeletclient.go
* Added network status annotations section to quickstart and added more unit tests
* made changes to tests based on code review
This fixes#243 with following changes:
+ Optimize to fetch Pod from k8s client
+ Change to use cache always in DEL.
+ If failed to fetch the pod info from k8s clinet in deletion,
use cached delegates as emergency bailout
+ Add test cases for cache
This fix is to add declaration about clusterNetwork/defaultNetwork
net-attach-def is in 'default' namesspace. In addition, this code
changes to skip defaultNetwork in case of 'kube-system' namespace
as well (#202).
When multus is run inside a container, eg as a daemonset, we can use
InClusterConfig() to automatically get the service account token, the
CA certificate, and the apiserver address, all without a kubeconfig
file.
Plugin went away, replaced by a minimal Spec.Config field so that the CNI
version the thick plugin supports could be indicated. This means we need
to also inject the network name into the minimal CNI JSON config when
loading the Spec.Config if the network name isn't there already.
