Making time off workable for an SRE team

The specialist is on vacation, but the service still knows their phone number. A backup is on the calendar, the pager reaches them and the runbook is linked. Then the recovery step asks for access only the absent person has. The handoff moved a name without moving the ability to do the job. Making time off workable starts by fixing that gap before anyone leaves.

Consider a hypothetical specialist whose vacation backup can acknowledge pages but cannot perform the recovery procedure. The useful rehearsal is to ask the backup to carry that procedure through with their own permissions while the specialist is still available. That turns a reassuring assignment on a calendar into a concrete question the team can answer.

Urgent work needs a route other than the absent specialist

Separate the urgent paging route from ordinary working-hours requests. Define what qualifies for the duty responder and where nonurgent work waits. That gives colleagues a practical alternative to privately contacting the person who usually knows the answer.

For the vacation backup, active investigations need more than a list of links. Transfer the current state, unresolved risks and the next consequential decision, including who can make it. That lets the backup take responsibility for the work instead of having to ask the absent specialist what the handoff meant.

Move responsibility before time away begins

Read diagram description

A coverage-planning sequence. Transfer or defer the work and check that the substitute can actually perform it. A name on the rota is not enough if essential access is missing. Diagram labels: Responsibilities during leave: Incidents, approvals and routine work; Transfer or defer: Identify backup coverage and work that can wait; Check the handoff: Access, knowledge and escalation path; Protected time away: The service plan no longer depends on the absent person.

What repeated private messages reveal

Repeated contact with one off-shift engineer can reveal a specific dependency: inaccessible tools, an undocumented recovery step, unclear decision authority or design knowledge that others have not had a chance to learn. Review what those messages actually ask for. A repeated request for a privileged command points to an access gap; repeated questions about a recovery step point to knowledge that has not transferred. Those are problems the team can address before the next absence.

A backup can discover missing access by attempting the recovery procedure while help is still available. Moving responsibility for the absence includes transferring the capability and authority that procedure requires. In the vacation case, rehearsing with the backup’s actual permissions shows whether the shift can finish the work without contacting the specialist.

The replacement also needs capacity. Review on-call load alongside rotation design so covering the specialist does not silently make the backup's own plan impossible. A sustainable transfer may require deferring routine work as well as distributing urgent responsibilities.

For example, suppose the backup can perform the recovery but already has a full project schedule. Transferring the pager has transferred a second workload, too. Decide which of the backup’s existing commitments moves before the absence begins. Otherwise the coverage plan may protect one person’s time away by consuming another person’s evenings.

Time away and recovery both change available capacity

A disrupted shift takes capacity from the next day as well as the night. If the same assignments and deadlines remain, the responder has to find those hours somewhere, often in the time that was supposed to be their own. Change the work plan with the recovery plan. The practical question is what the available people can now fulfill.

Recurring interruptions deserve the same accounting. Depending on their source, a team may need to narrow service scope, change support intake, fund automation or add trained coverage. These choices address the relationship between demand and capacity rather than asking an individual to organize an impossible calendar more carefully.

Agreed channels and recovery arrangements need to remain dependable when delivery pressure rises. Leaders following them helps colleagues plan around the arrangement instead of guessing when it will be overridden. The planning conversation can stay focused on coverage and available capacity; people need not disclose unrelated personal circumstances to justify a workable schedule.

Exceptional emergencies may exceed ordinary coverage. Define that exception narrowly and account for the resulting work afterward. Repeated reliance on the exception is evidence to repair the coverage model rather than a reason to assume the absent person is routinely available.

A rehearsal can reveal the missing access while there is still time to grant it, choose a different backup or reduce the commitment. The covering engineer knows what they can do and where help will come from. The specialist can leave without remaining the plan’s invisible final step. That is a more dependable basis for time off than asking either person to be better at ignoring a phone.