Skip to main content
Question

Query about middle tier pod resources

  • December 16, 2025
  • 2 replies
  • 13 views
G
Do Gooder (Customer)
Forum|alt.badge.img+1

Hello, 

We in the process of upgrading from Apps 10 UPD24 to Cloud 24R2 SU8. 

Our test environment has been configured with a global scale value of 20 in ifscloud-values.yaml. 

We had issues with the ifsapp-odata pod being killed due to hitting memory limits (OOMKilled message in Linux journal logs), our upgrade partner then added this to ifscloud-values.yaml to prevent this from happening:

 

ifsappodata:
  replicas: 1
  resources:
    requests:
      cpu: "500m"
      memory: "5Gi"
    limits:
      cpu: "2"
      memory: "5Gi"

 

Do IFS recommend specifying pod resources this way? 

Our MT server is at 18GB out of 96GB of memory used - would it not be better to increase the scale value/add more replicas for ifsapp-odata?

 

Thanks, 

Joe

2 replies

H
Hero (Employee)
Forum|alt.badge.img+11
  • hhanse
  • Hero (Employee)
  • December 16, 2025

This is how Scale is defined:

Dev might have as low as 10-20  (depending on workload size - one big transaction can cause OOM)
Test is 50-100 (depending on workload - one big transaction can cause OOM)
Prod is always 100.

G
Do Gooder (Customer)
Forum|alt.badge.img+1
  • Galvanize96
  • Author
  • Do Gooder (Customer)
  • December 16, 2025

Hi Hhanse, 

 

Thanks for responding - do you know if configuring per pod limits like above is recommended? Or should we be increasing the scale/number of replicas as a first step?

 

Thanks, 

Joe 

 

Terms & ConditionsCookie settingsAccessibility statement