The local node where you initiate the command must remain up (the GI stack must be running).
Applying patch 72030 on node racnode1... (non‑rolling mode) Applying patch 72030 on node racnode2... OPATCHAUTO succeeded. opatchauto72030 execute in nonrolling mode
If you must apply it non-rolling (e.g., during a complete blackout window), ensure all services are stopped and use the override if available, though this is generally discouraged for RU patches. 4. Update OPatch and OPatchAuto The local node where you initiate the command
| Issue | Action | |--------|--------| | PRIF-10XX (OCR errors) | Run ocrcheck , ensure voting disks online | | opatchauto apply -nonrolling fails mid‑way | Use opatchauto resume after fixing cause | | Inconsistent patch state across nodes | Run opatchauto rollback /patch/path -nonrolling | OPATCHAUTO succeeded
. In a shared home configuration (like those found on some Cluster File Systems), you cannot patch one node while the others are still running off the same binaries. This creates an invalid execution mode, triggering the dreaded OPATCHAUTO-72030 The Turning Point: Making the Hard Choice
The number 72030 typically refers to a specific Oracle patch UID. In internal testing or legacy documentation, you might see this used as a placeholder. In practice, your command will look like this:
If opatchauto continues to fail in non-rolling mode, you can fall back to the manual opatch method: Stop the stack on the node.