How to Write a Smart Hands Ticket / MOP So Remote Techs Don't Fail You
A good smart hands ticket tells a technician exactly which rack, which device, which steps, how to verify success, and when to stop and call you. A weak ticket produces the classic Reddit complaint: remote hands refuse the work, or botch it, and you fly out anyway.
This is the practical companion to what smart hands means and smart vs remote hands.
On this page
Minimum fields every ticket needs
- Facility, suite/cage/rack, and asset labels or serials
- Change window and maximum acceptable impact
- Named remote contact with phone + authority to approve
- Step-by-step MOP with success criteria after each critical step
- Rollback or stop conditions
- Photos or diagrams for cabling and port maps when relevant
MOP structure that works on the floor
- Pre-checks: power, link lights, current state photos
- Action steps numbered and one action each
- Verify: command output, LED state, or remote console confirmation
- Document: time, serials, photo of finished state
- Close: confirm with remote owner before leaving the rack
Before the ticket: access and notice
- Access request: the engineer cannot get past security without it. With eWEB TEC, the customer either lodges the work visit and adds our engineer to their access list, or we raise it on the customer’s account with written authority.
- Notice: 24 hours for a scheduled change window. Faults are dispatched 24/7.
- Window: business hours are 8am–6pm weekdays. Anything outside that is after-hours.
- Budget: each call-out carries a 2-hour minimum, so group small tasks for one rack into one ticket. See what smart hands costs.
Common failure modes
- Vague labels ("the top switch")
- No serial or port map for cable moves
- Assuming the tech has your VPN or passwords they should not need
- No bridge contact overnight
- Scope creep mid-ticket without a written add
Sydney facilities note
Access rules differ across Equinix, NextDC, Global Switch and others. Include escort/access method and any site-specific rules you already know. Orientation: Sydney data centres guide. Service: smart hands.
FAQ
What is a MOP in a data centre?
A Method of Procedure is a step-by-step change plan: what to do, in what order, with what success checks and rollback.
Do simple reboots need a full MOP?
Even short tickets need clear asset ID, authority to act, and a verify step. Complex installs always need a written MOP.
Who should stay on the bridge?
Someone who can approve escalations and confirm success criteria while hands are on the floor.
How much notice does eWEB TEC need for a planned change?
24 hours for a scheduled change window, so the facility access request can clear. Faults and outages are dispatched 24/7 without notice.
Who raises the data-centre access request?
Either. The customer lodges the work visit and adds our engineer to their access list, or we raise it on the customer’s account with their written authority. Which one depends on the customer and the facility.
Need hands on the floor with a clean MOP?