add feature HPA on core, trivy and registry

[flbla]

Hello,
It could be great to have HPA (horizontal pod autoscaling) option on core, trivy and registry.

In my case I have few load during night, but much more during day as dev teams are building images and doing a lot of deployments.

Thanks

[ywk253100]

We need to do some investigation for the HPA first

[flbla]

@ywk253100 : any ETA for this feature ?

[Tejuvmware]

Hello @ywk253100 , ywk253100 ,
It's the same request that i wanted to come up with. I'm also facing the same issue and need HPA enabled for the registry and core pods. any ETA on this feature ? or Is there any alternative method to apply HPA to the clusters ?

Please let me know as soon as possible because we are expecting a huge traffic with an application release and need HPA to be implemented ASAP.

[pascalrimann]

is there any information about the HPA? Would be nice to have it

[yashid-mohamed]

We are in the same boat and would like to automatically scale up based on demand (example during scheduled scans of all images)

[plodocus]

We need to do some investigation for the HPA first

We also have this requirement. What kind of investigations need to be done?

How have people in this thread mitigated their issues? Did you add your own HPAs?

[flbla]

How have people in this thread mitigated their issues? Did you add your own HPAs?
yes, I created my own

[rekha-prakash-maersk]

@flbla Can you please shed some light on your HPA setup ? we are using helm chart deployment for our harbor setup, and looking to implement a HPA for Trivy based on the load.

[flbla]

it depends of your need
I choosed to do it with external metrics based on the harbor_task_queue_size/image_scan metric.
it's a metric exposed by the harbor exporter.

[xoanmi]

I've added HPA and Kepa autoscaling on this PR: #1756