* Define and use an interface for logging.
This allows alternative implementations to use whatever logging package
they want.
Currently the interface just mimicks what logrus already implements,
with minor modifications to not rely on logrus itself. I think the
interface is pretty solid in terms of logging implementations being able
to do what they need to.
* Make tracing interface to coalesce logging/tracing
Allows us to share data between the tracer and the logger so we can
simplify log/trace handling wher we generally want data to go both
places.
This ensures that we can catch certain types of errors from providers
and handle accordingly in the core. There is still more to do here to
improve that but this resolves an immediate need to know why a Delete
failed.
vic provider was not updated since I could not figure out where to get
this information.
NRP used to add a profile entry when a service association link is added to the subnet
Now, the profile entry is added when a network profile is created
This breaks the current validation
* Add default provider taint and taint configuration options
This allows for more specificity when setting taint tolerations for
workloads. Three new env variables are introduced:
VKUBELET_TAINT_KEY (defaults to `virtual-kubelet.io/provider`)
VKUBELET_TAINT_VALUE (defaults to provider name)
VKUBELET_TAINT_EFFECT (defaults to `NoSchedule`)
BREAKING CHANGES:
- The default taint key of `azure.com/aci` is now
`virtual-kubelet.io/provider`.
- Specifying a custom taint key is now done via an environment variable
rather than the `--taint` command line flag.
* Add back deprecated taint flag
TODO: Revert this commit
Related to #316
* Add darwin tag to not build for osx
* Darwin specific lookup file without cri and vic
* Fix chart notes template (#317)
Values were moved from env to top level.
* If --taint is specified, set the taint value to empty (#322)
Add the old tolerations the examples to make it backward compatible during the switch
* Use standard logging package (#323)
* Update kubelet vendor to pull in stats API
* Add errgroup dep which will be used for ACI stats
* Add supports for stats in ACI provider
This adds a new, optional, interface for providers that want to provide
stats.
* Don't use globals for API server
Refactors how HTTP servers are started and binds them to objects that
can store the provider rather than relying on a global.
* Fix merge conflict
* Fix couple errors
* Update vendor for azure vent support
* Add support for Azure custom vnets.
Use pointers intead of values. This allows the client to pass back
returned data from Azure.
Update k8s client and the dependencies
ACI client change for Mocking
Add ACI Provider Mock Tests
Add the Windows development environment
Add UT for Default Resource Requests
Enable the make test in Docker file
Update the vendors
* Read ACS Credentials for Azure Authentication
Supprt a new environment variable: ACS_CREDENTIAL_LOCATION
Expect the value to be the ACS credential filepath, which is the
/etc/kubernetes/azure.json file generated on the ACS nodes.
If the ACS_CREDENTIAL_LOCATION is specified and loaded,
create the Azure Authentication class from its values.
If the AZURE_AUTHENTICATION_LOCATION is specified and loaded,
its values will overwrite the value above.
Refactor the ACI provider and ACI client to be able to override the SPN by environment variable
