Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Cron Scaler to Scale the deployments as per a particular cron schedule. #812

Closed
ckuduvalli opened this issue May 11, 2020 · 7 comments · Fixed by #858
Closed

Cron Scaler to Scale the deployments as per a particular cron schedule. #812

ckuduvalli opened this issue May 11, 2020 · 7 comments · Fixed by #858

Comments

@ckuduvalli
Copy link
Contributor

ckuduvalli commented May 11, 2020

A clear and concise description of what scaler you'd like to use and how you'd want to use it:

Cron scaler is being developed in order to scale the deployments based on a particular time schedule. Whenever the currentTimestamp is inbetween a start and endTime, we say the scaler is active and inactive in all other times.

  • Scaler Source: Deployment replicas from the k8s deployment in the namespace itself.
  • How do you want to scale: The cron scaler will scale the pods to the desiredReplicas when the scaler is active. By Active, I mean the timestamp should be between the start and end time specified in the metadata. All the other times, the replicas will be equal to the minReplicaCount
  • Authentication: None
@ckuduvalli ckuduvalli changed the title New Scaler to Scale the deployments as per a particular cron schedule. Cron Scaler to Scale the deployments as per a particular cron schedule. May 11, 2020
@ckuduvalli
Copy link
Contributor Author

By scaler source, I understand something like kafka topic, rabbitmq etc. Please correct me if I am wrong.

@zroubalik
Copy link
Member

@ckuduvalli yeah scaler source, is the source of the events, as you mentioned eg. kafka topic..)
In this case, I suppose the source would be the cron implementation itself?

There's community call this Thursday https://keda.sh/community/, you can join and we can discuss it.

@ckuduvalli
Copy link
Contributor Author

ckuduvalli commented May 12, 2020

@zroubalik I have a question regarding my scaler. I implemented a cron scaler. What I am observing is that keda passes the metric to scale up to some value x, but HPA is scaling to some value less than x due to CPU utilization taking over. Is there a way I can have my scaler override the HPA?

As per the documentation, I see that HPA should scale the replicas to max of the 2 metrics
https://kubernetes.io/docs/tasks/run-application/horizontal-pod-autoscale/#support-for-multiple-metrics

@ckuduvalli
Copy link
Contributor Author

@ckuduvalli yeah scaler source, is the source of the events, as you mentioned eg. kafka topic..)
In this case, I suppose the source would be the cron implementation itself?

There's community call this Thursday https://keda.sh/community/, you can join and we can discuss it.

Sure.. I will join the call on Thursday

@balchua
Copy link
Contributor

balchua commented May 12, 2020

This is interesting and at the same time it doesn't need hpa to determine how many to scale to. Similar to ScaledJob.

@ckuduvalli
Copy link
Contributor Author

This is interesting and at the same time it doesn't need hpa to determine how many to scale to. Similar to ScaledJob.

This happened due to my misunderstanding of the metric. Will discuss over the call today.

@tomkerkhove
Copy link
Member

Good idea for a scaler if you ask me!

@tomkerkhove tomkerkhove added this to the v1.5 milestone Jul 7, 2020
SpiritZhou pushed a commit to SpiritZhou/keda that referenced this issue Jul 18, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging a pull request may close this issue.

4 participants