Recommendation UI shows recommendation service unreachable error message banner because underlying Recommendation Pod OOM when large number of recommendations are in the system

Products

• VMware vDefend Firewall
• VMware vDefend Firewall with Advanced Threat Prevention

Issue/Introduction

Recommendation Pod was observed to restart after running for 22 hrs with below config
- Deploy setup with 10 WN
- Enable intelligence,NTA,MPS,NDR,Rule analysis
- Scaleout all verticals.
- Bootstrap CSV which will create 2500 applications,36k VMs
- Flow generation  at 1M/5mins

Environment

SSP 5.1.1

Cause

Happens only when lot of recommendations and Apps are present in the system. When lot of recommendations are in the system, the default memory of the recommendation pod of 2G is sometimes not sufficient to load all the recommendation summaries in the UI. 

k describe pod recommendation-<xxxx>-<yyyy>

You should see something like below with Exit Code: 137 and Restart Count: <greater than 0>
This shows that the Recommendation Pod was killed by the Kubernetes Control Plane due to exceeding 2Gi memory limits set forth on it as shown below

Last State: Terminated
      Reason: OOMKilled
      Exit Code: 137
      Started: Sat, 10 Jan 2026 01:40:53 -0800
      Finished: Thu, 15 Jan 2026 07:45:47 -0800
    Ready: True
    Restart Count: 2
    Limits:
      cpu: 8
      ephemeral-storage: 1Gi
      memory: 2Gi

Resolution

Resolution requires giving more memory to recommendation pod; This requires below steps

k edit deployment recommendation

Find the below section which specifies the memory
Change the memory to 2560Mi i.e. 2.5G 
Save the edited configuration by using command :wq
A new recommendation pod should come up and recommendation UI should no longer show error messages.

If the problem persists, repeat the above steps to Change the memory to 3Gi

Limits:
      cpu: 8
      ephemeral-storage: 1Gi
      memory: 2Gi