Dell AIops: Deployment of Long-Running Workload Blueprints Fail with Memory Error

Long-running deployments fail.

Sidecar containers (hzp_iam_proxy) indicate an Out of Memory error. The following is an example:

  - containerID: containerd://63054fd6b999e08cc13add411b96e40bfd28d197e0008deff7a4e13d62411218
    image: 172.28.177.1:5000/hzp/hzp-iam-proxy:1.2.0.0-204fa20-f30c4736
    imageID: 172.28.177.1:5000/hzp/hzp-iam proxy@sha256:98d808392f2e67e7bd56d7e55b788f8fd3c3f152cc38051dcd9003eec121bdab
    lastState:
      terminated:
        containerID: containerd://ccb2e7ca18aabe8bc3a813893f2f885c66323974a51dba63bd9b0a699fcf8a21
        exitCode: 137
        finishedAt: "2026-04-21T16:10:57Z"
        reason: OOMKilled
        startedAt: "2026-04-10T17:53:05Z"

The Dell Automation Platform Helm Plugin version 1.4.3 and later uses more memory than expected in the "fusion-api container" and in the sidecar within the "hzp-eoproxy svc container."

1. Increase the memory limit in the "fusion-api service."
   kubectl edit deploy -n dapo fusion-api-service

   When the deployment editor opens, find the resources section under the main container (api-service). Edit the "limits.memory" value.

   Change the value from 512Mi to 1000Mi:

   resources:
     limits:
       cpu: "1"
       memory: 512Mi      # <-- change this value
     requests:
       cpu: 100m
       ephemeral-storage: 100Mi
       memory: 128Mi

   Note: There are two resource blocks in this deployment. The correct block to edit is the "api-service container," not the "api-service-init container."

   Example of change applied:
   

2. Increase the stateful set of "hzp-eoproxy-svc."
   kubectl edit sts -n dapo hzp-eoproxy-svc

   serviceName: eoproxy-headless  
   template:
       metadata:
         annotations:
           hzp.iam.webhook/authentication-enabled: "true"
           hzp.iam.webhook/authorization-enabled: "true"
           hzp.iam.webhook/resources-limits-memory: 1000Mi # <-- add this line

   Example of applying the change for the "hzp-eoproxy-svc":
   

This issue is planned to be addressed in a future release of Dell Automation Platform.

If this issue occurs on a deployment in a Dell Automation Platform as a Service environment, contact Dell Support to engage Engineering. Reference this article ID when contacting support. An example of the deployment failure is helpful to determine whether the issue matches this issue.