My employer canceled the support contract for a mission-critical production system but still expects me to perform an update. Management says the change is straightforward and that the vendor rarely performs this update for customers. I'm concerned about the risk, especially if something fails and we can't get vendor assistance. What steps would you take before proceeding, and how would you protect the system and document the decision?
4 Answers
Vendor support isn’t always required for routine maintenance, but you need to understand the product and follow its documentation carefully. Treat it like any other production change: back up what you can, capture a VM checkpoint if appropriate, schedule downtime, validate the result, and be ready to roll back immediately.
The important part is having a recovery plan, not just hoping the update works. Confirm that the backup can actually be restored, define who makes the rollback decision, and reserve enough time to recover before normal operations resume. If management refuses testing, backups, or rollback time, that’s a much bigger operational risk and should be escalated.
Document the risks and your concerns in writing, including the lack of vendor support, and ask management to explicitly approve proceeding. Put the change through your normal change-control process, identify the maintenance window and affected services, and make sure everyone knows what happens if the update fails.
Test the update on a development environment, clone, or other non-production system first. Make sure it behaves correctly and that any important integrations still work. Before touching production, verify that you have a known-good backup or snapshot and a tested rollback procedure.
That makes sense. I’ll send a written summary with the risks, proposed steps, and a clear request for approval rather than relying on a verbal go-ahead.