* Drop MaxWeightedCollector (we don't want max anyway, we want sum)
* Use Prometheus to add up all matching metrics and scale them; this
has a nice side effect of ensuring that unused hostnames don't cause
an error when collecting the metrics
* Update the tests a bit
Signed-off-by: Alexey Ermakov <alexey.ermakov@zalando.de>
This adds support for `averageValue` for the `request-per-second` metric
based on Ingress Objects. This is only supported from Kubernetes
`>=v1.14` (https://github.com/kubernetes/kubernetes/pull/72872).
When defining the HPA with `autoscaling/v2beta1` you still need to
define `targetValue` even though it won't be used when `averageValue` is
set. Once we default to `autoscaling/v2beta2` this akward API will be
gone.
Signed-off-by: Mikkel Oscar Lyderik Larsen <mikkel.larsen@zalando.de>
* Added weighting of rps metrics based on backend weights
Signed-off-by: Arjun Naik <arjun.rn@gmail.com>
* Updated documented with instructions on how to use the backend weighting
Signed-off-by: Arjun Naik <arjun.rn@gmail.com>
* Changed separator for RPS metric and added flag to specify backend weights annotation.
Signed-off-by: Arjun Naik <arjun.rn@gmail.com>
* Allow for multiple backends with for weighting.
Signed-off-by: Arjun Naik <arjun.rn@gmail.com>
* Look at total replicas instead of readyReplicas
In cases where replicas take a bit of time to get ready, the Stackset Controller just keeps on scaling. To avoid this issue,
using the total replicas instead of replicas that are ready.
Signed-off-by: Muhammad Muaaz Saleem <muhammad.muaaz.saleem@zalando.de>
* Fix bitbucket clones
Signed-off-by: Alexey Ermakov <alexey.ermakov@zalando.de>
* Updating .gitignore
Signed-off-by: Muhammad Muaaz Saleem <muhammad.muaaz.saleem@zalando.de>
* Fixing Typo
Signed-off-by: Muhammad Muaaz Saleem <muhammad.muaaz.saleem@zalando.de>
* Adding tests to avoid regression
Testing that collectors depend on Resource.Status.Replicas and not
Resource.Status.ReadyReplicas
Signed-off-by: Muhammad Muaaz Saleem <muhammad.muaaz.saleem@zalando.de>
