1. PostgreSQL HA Meeting Summary (Oct 28, 2025)
    
   1. Core Decision
      ACM will consolidate 4 separate PostgreSQL instances into 1 for 2.16 release. This is the immediate priority, regardless of HA strategy.
       
      —
       
   2. The Problem

• {}Kessel ABAC{} needs HA PostgreSQL for production on-premises MCE
• {}ACM cannot build HA internally{} (requires 3 dev FTEs + 1 maintenance FTE)
• {}Not ACM's responsibility{} - Red Hat is forming cross-BU community (Quay, ACS, Ansible) to address
   
  —
   
  1. Red Hat's Commitment
     {}Batteries Included Model:{}
• Red Hat will provide supported, complete product (NOT BYODB)
• Currently: Single non-HA PostgreSQL instance
• Why: 90% of market lacks PostgreSQL expertise; only "whale" accounts can do BYODB
   
  —
   
  1. 2.16 Release Plan
      

     Item	Decision
     ------	----------
     Consolidate 4 PostgreSQL instances	Priority for 2.16
     Kessel Tech Preview	Possible with consolidated instance
     HA Required for Preview	NO
     Existing ABAC Solution	Will go GA (CMV pressure)
     HA Required for Production Kessel	YES (long-term)

 
—
 

1. 1. PostgreSQL Consolidation Priority
      {}Current 4 separate instances:{}
      1. Search
      2. Global Hub
      3. Assisted Installer
      4. Policy Compliance History
       
      {}Goal:{} Single consolidated instance/image for efficiency and resource savings
       
      —
       
   2. Action Items
      1. Read Quay's PostgreSQL HA document
      2. Discuss PostgreSQL HA with Quay team in OP meeting
      3. Update ACM-11471 epic with meeting notes
       
      —
       
   3. Key Context

• Most customers run <10 clusters and lack PostgreSQL expertise
• Current non-HA setup is known risk but hasn't significantly impacted business yet
• Previous Crunchy partnership attempts didn't materialize
• Cross-BU collaboration expected with Junis James Lasco's team as key contributor