"No connection pool for 'ActiveRecord::Base' found" returned on OpsMan upgrade during import

Products

• VMware Tanzu Kubernetes Grid Integrated Edition
• VMware Tanzu Kubernetes Grid Integrated (TKGi)

Issue/Introduction

• During upgrade of Operations Manager, while Importing the OM backed up data into the new VM, users might see the following error returned:
  
  No connection pool for 'ActiveRecord::Base' found
  
  
• The error stops OpsMan upgrade by preventing completion of the "Import Existing Installation" operation on the new OpsMan VM
• Viewing the /var/log/opsman/production.log on the new OpsManager after import failure, users might see errors like:
  

  Started GET "/api/v0/info" for IP_ADDRESS
  No connection pool for 'ActiveRecord::Base' found.
  Exception caught via rescue from (code 500):

• This error might appear intermittently

Environment

Operations Manager versions prior to 3.0.23

Cause

This error is caused by external servers or users attempting to access the /api/v0/info endpoint on the new OpsMan VM, interrupting the import process.

Resolution

This problem is reported as resolved in the OpsMan v3.0.23+LTS-T release notes

 

Workaround:

 

If upgrade is not an option, potential workarounds include:

 

1. 1. Identify the server causing the 500 errors by reviewing the /var/log/opsman/production.log. The reported 500 error message should include an IP address.
   2. Identify which server the IP address is associated with and temporarily stop it from connecting to the OpsMan VM.
   3. If necessary, the following commands can be used from the new OpsMan server directly to temporarily block the offending IP from interrupting the import process:
      
      

      iptables -I INPUT -s OFFENDING_IP_ADDRESS -j DROP
      
      

   4. Retry the Import steps again.
   5. Ensure the iptables rule is removed after this work to allow access to the OpsMan server.