Error Adding Projects to Kubernetes Zone: Exceeded Maximum Buffer Size Limit
search cancel

Error Adding Projects to Kubernetes Zone: Exceeded Maximum Buffer Size Limit

book

Article ID: 384585

calendar_today

Updated On:

Products

VMware Aria Suite

Issue/Introduction

  • An error is encountered while adding projects to a Kubernetes zone in Aria Automation version 8.16.2. Specifically, when attempting to create the 994th project, the system returns the error:
    • "Some of your changes could not be applied: Exceeded limit on max bytes to buffer : 262144."

  • This article provides details about the issue, its cause, a workaround, and the planned resolution

Environment

  • Aria Automation 8.16.2
  • Aria Automation 8.x

Cause

  • The issue arises due to a buffer size limitation (262144 bytes) in the system when handling a large number of projects. This buffer size is exceeded during the creation of additional projects, preventing the successful addition of new projects beyond the current limit

Resolution

Workaround:

  • The following workaround can be applied to address the issue and allow for the addition of more projects:

 

  • Important Notes:
    • The workaround involves downtime, as all services will be redeployed.
    • Ensure simultaneous VM snapshots (without memory) of all Aria Automation nodes are taken before proceeding.

 

Steps:

1. Take Snapshots:
   i. Take simultaneous VM snapshots without memory of all three Aria Automation nodes (if applicable). Snapshots can be taken via Aria Suite Lifecycle or manually from vCenter.

2. Create the Custom Profile: 
   i. Execute the following command on one of the Aria Automation nodes to create a custom profile:

vracli cluster exec -- bash -c 'base64 -d <<< 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 | bash -'

3. Deploy the Custom Profile Configuration:
   i. Execute the following script on one Aria Automation node:

/opt/scripts/deploy.sh

4. Removing the Custom Profile (Optional):
   i. If the configuration is no longer required, remove the custom profile directory from all nodes using the command:

vracli cluster exec -- bash -c 'rm -rf /etc/vmware-prelude/profiles/cmx-service-increase-spring-codec-memory'

  ii. Reapply the deployment script from step 3 after removing the profile.

 

  • Note: By applying the above workaround, users can add approximately 4500 zones to a project without encountering the error.

 

Fix in Future Release:

  • A permanent fix for this issue is planned for inclusion in Aria Automation 8.18.1 Patch 2.

Additional Information

  • Ensure proper testing and validation are performed after applying the workaround