# 2026-04-30 | Replica Pipeline for CI DB failed (Severity 4) **Problem**: The CI database replica pipeline failed, first because the target VM was terminated, and again after a restart due to a data disk size mismatch with the required snapshot. **Impact**: There is no customer impact or production degradation. The CI database replica pipeline could not complete, which may have delayed or blocked updates dependent on the replica, but no production deployments or feature flags were affected. **Causes**: The initial failure was caused by the target VM being in a terminated state. After restarting the VM, the replica pipeline failed again because it tried to create a data disk of 10,000GB, which was too small for the required 16,000GB snapshot, resulting in a GCP disk creation error. Later found that source disk/server for patroni CI is no longer available and needs an update. **Response strategy**: We increased the data disk size to 16,000GB and merged the fix. The pipeline has been re-run from main to test the changes, and the rebuild process is underway. A separate improvement to add a preflight check for VM status is still pending approval. Also the source data disk for patroni-ci has been updated to use updated patroni-ci cluster nodes. _This ticket was created to track_ [_INC-9700_](https://app.incident.io/gitlab/incidents/9700)_, by_ [_incident.io_](https://app.incident.io) 🔥