Hello, our infrastructure team is trying to figure...
# gooddata-cnp
Pete Lorenz
07/12/2023, 3:20 PMHello, our infrastructure team is trying to figure out how much capacity is needed for the Pulsar and GoodData services. They want us to provide cpu and memory usage requirements and number of instances required for each service to provision instances of the optimal size. Would anyone know where I can obtain this information?
r
Robert Moucha
07/12/2023, 3:41 PMI will more my previous answers to this thread to keep them related to your question.
Robert Moucha
07/12/2023, 3:41 PMThe recommended settings are here
Robert Moucha
07/12/2023, 3:41 PMBut we're preparing update of this page where we will describe how to install Pulsar chart 3.0.0. The values had changed significantly
Docs are not yet published (we're waiting for upcoming 2.4.0 release of CN)
I can send you updated values for pulsar 3.0.0 chart.
The major benefit of new pulsar chart is it works on k8s 1.25
👍 1
p
Pete Lorenz
07/12/2023, 3:43 PMthank you Robert 😀
r
Robert Moucha
07/12/2023, 3:46 PMThis is values.yaml file for installing apache/pulsar v 3.0.0. If you want to upgrade from version 2.9.x you MUST enable the
securityContext.fsGroupChangePolicy👍 1
Robert Moucha
07/12/2023, 3:50 PMRegarding the resource limits, they are not set in that file. These values should be sufficient for the most deployments:
Copy code
---
zookeeper:
resources:
requests:
memory: 512Mi
cpu: 100m
limits:
memory: 512Mi
cpu: 100m
bookkeeper:
resources:
requests:
cpu: 200m
memory: 512Mi
limits:
cpu: 200m
memory: 512Mi
broker:
resources:
requests:
cpu: 200m
memory: 640Mi
limits:
cpu: 200m
memory: 640Mi
autorecovery:
resources:
requests:
memory: 256Mi
cpu: 100m
limits:
memory: 256Mi
cpu: 100m👍 1