Dedicated Migration Team
[[TOC]]
Team Mission
Partner with the field, customers, and engineering to deliver seamless, fast, and reliable migrations. The team executes and improves processes to eliminate uncertainty and ensures readiness for cutover in Geo-based Dedicated migrations.
Priorities
- Deliver and refine toolkits that include the documentation, templates, talk-tracks, and tooling to smoother Geo Dedicated migrations for ourselves and our customers
- Work with our engineering peers on features that enable self-serve customer migrations
- Deliver in-flight Geo migration engagements
- Enable our PS peers to deliver Geo migration engagements to scale up and out
- Delivery tooling and process to enable faster scoping and sizing for the field in the pre-sales process
Team members
| Name | Role |
|---|---|
| Eren Akca | Staff Migration Engineer |
| Petar Prokić | Staff Migration Engineer |
| Jessykah Bird | Senior Migration Engineer |
| Glen Miller | Senior Manager, Software Engineering |
Roles & Responsibilities
The team:
During the Geo migration portion of a Dedicated engagement, the team has the following responsibilities:
- Works directly with the field and customers to produce automated solutions for technical discovery, validation of data, and capacity planning of their target instance
- Owns communications, providing internal and external alignment of plans, status, and timings
- Owns Project Management alignment with the customer Project Planning Office for clear synchronization of activities, comms, timelines, and responsibilities
- Enables confident production switchovers with minimal risk and maximum speed
Operating Principles
Customer-Centric Excellence
- Clear Expectations First: Every engagement begins with explicit alignment on migration cadence, platform capabilities, and service model boundaries
- Consultative Partnership: We act as trusted advisors, not order-takers, providing honest guidance even when difficult
- Transparent Communication: Regular, structured updates with no surprises - customers always know where they stand
Technical Rigor
- Discovery Before Design: Comprehensive technical and business discovery drives all solution approaches
- Validation at Every Step: Multi-layered validation ensures accuracy before any production changes
- Automation Where Possible: Standardized, repeatable processes reduce risk and improve consistency
Structured Delivery
- Milestone-Driven Progress: Clear checkpoints with customer sign-off prevent scope creep and misalignment
- Risk-First Planning: Identify and mitigate risks early rather than hope for the best
- Documentation as Code: Every process, decision, and artifact is documented for repeatability and knowledge transfer
Continuous Improvement
- Learning Organization: Every project contributes to improved processes, tools, and methodologies
- Cross-Functional Collaboration: Regular feedback loops with Engineering, Product, and Customer Success teams
- Scalable Solutions: Build capabilities that enable future self-service and automation
Internal Alignment
- One Team Approach: Clear role definitions with flexible collaboration across disciplines
- Shared Accountability: Success is measured at the team level, not individual contribution
- Proactive Communication: Internal stakeholders are informed and aligned before customer commitments
Success Metrics
- To be aligned with larger team metrics in progress
Process
Engaging with Others
Geo Engagement Model For Dedicated Migrations
Pre-production Cutover and Requests Outside the 2-Week Production Support Window
For support during any period outside the production cutover window, please file an RFH (see below). This include pre-production cutover. There will not be an assigned Geo engineer for pre-production. Customer call requests are handled adhoc on a case-by-case basis, requiring approval from Engineering Manager Lucie Zhao based on team availability.
Production Cutover and Production Support Window
A Geo DRI is assigned for a 2-week window around each production migration: 1 week before, the migration weekend, and 1 week after. The DRI is fully committed to migration issues via Slack during pre/post weeks and available via PagerDuty during business hours (9am-5pm in their timezone) on the migration weekend itself. If no issues arise, the DRI can resume normal Geo duties.
Migration Weekend Responsibilities
The Geo team must have someone prepared and aware of the cutover, coordinate timing with the Project Manager, and be ready to troubleshoot any issues that arise during the customer instance cutover following full data sync.
RFH Process
For sync issues outside the production cutover window, teams submit a Geo Support Request indicating priority (P1-P3), sync start time, cutover deadline, tenant, and request. These are triaged by the weekly customer support DRI. Geo engineers need access to Opensearch, Grafana, and ideally break glass access to support effectively.
| Priority | Description |
|---|---|
| 1 | Immediate blocker to data sync or pre-production cutover. Cannot proceed without a fix |
| 2 | Needs to be solved before the end of the pre-production cutover phase, but not a blocker to start |
| 3 | Needs to be solved before production cutover. Can be handled within the production cutover support window |
SRE Engagement Model for Dedicated Migrations
Pre-production Cutover
A rotating team of SREs is assigned to the pre-production cutover. This is generally the same team that will support the production cutover. Pre-production cutover is often a more measured process that often runs for multiple days. It is where we map out any potential issues with the production cutover. Teams are not required to be on a call bridge. Execution, communication, and hand-overs can be coordinated in #g_dedicated-migrations channel and in the issues for the engagements in the Dedicated Migrations project
Production Cutover and Production Support Window
A team of SREs is assigned for a 2-week window around each migration: 1 week before, the migration weekend, and 1 week after. This is generally the same team that supported the pre-production cutover. The assigned SREs are to treat all migrations issues as Priority 1 during pre/post weeks. Communication can occur in the #g_dedicated-migrations channel and in the issues for the engagements in the Dedicated Migrations project
Migration Weekend Responsibilities
The assigned SREs are fully committed to migration issues via Slack during pre/post weeks and will alternate duties during cutover weekends based on their location/timezone. The SRE team will be responsible for executing scripts outside the permission framework of the Dedicated Migrations team, and assisting in troubleshooting of infrastructure (networking, nodes, configuration) concerns.
Note
Many customers need to perform “catch-up” GitLab upgrades in the post-support window. The Dedicated Migrations team should be responsible for coordinating with the customer, scheduling, and executing on these upgrades, with support from the SRE team.
Outside the 2-Week Window
Support is primarily asynchronous via Request for Help (RFH) tickets. Per @fviegas all Dedicated Migrations RFH are considered priority 1. Synchronous customer calls are handled adhoc on a case-by-case basis, requiring approval from one of Steve Denham/Oriol Lluch Parellada based on team availability.
RFH Process
For sync issues outside the dedicated window, teams submit a Dedicated Support Request indicating the tenant and request.
Scheduling
SRE and Geo Coverage
The engineering managers will coordinate finding the proper resources for cutover windows. When PMs have locked-in dates and times for cutovers, and need coverage confirmation, they can reach out to @glen_miller to coordinate with the other EMs. The EMs will use a single epic to track the coverage for each cutover slot.
Cutover Slots
Geo migration cutover slots are currently a limited resource. Available slots can be referenced here. Establishing these slots should be part of the pre-sales process. Work with your Dedicated PMs and EMs to reserve a slot.
Scheduling Cutover Slots
When scheduling cutover slots, the following guidelines apply:
- Cutover slots are scheduled on a first-come, first-served basis
- Slots are typically available on weekends to minimize customer impact
- Slots should be set at time of deal close. If not, they should be confirmed no later than 14 weeks before the slot date.
- Slots are confirmed only after SRE and Geo coverage is confirmed by the respective EMs
Dedicated Migration Team Time Off
The Dedicated Migration team follows GitLab’s paid time off policy. We balance that with the need to deliver support to our customers every day. This page is intended to provide an understanding of what we need to do to achieve that balance, making it possible for all to take time off as needed and desired, while we as a team continue to deliver for our customers.
One-time setup actions
Dedicated Migration Team calendar
-
Be sure you have access to the Dedicated Migrations Team Calendar team calendar.
- If you don’t have it, verify whether you are part of the Dedicated Migrations Team (
dedicated-migrations-team@) if you are not, ask to be added via an Access Request.
- If you don’t have it, verify whether you are part of the Dedicated Migrations Team (
-
Set up the Google Calendar integration with Time Off by Deel, so that you do not need to populate your personal and “Dedicated Migrations Team” calendars manually.
- In Slack, click the
+sign next to ‘Apps’ at the bottom of the left sidebar - Search for ‘Time Off by Deel’ and click ‘View’
- Under ‘Home’, click on ‘Your Events’ to show a dropdown
- Click on ‘Calendar Sync’ under the Settings break
- Click on ‘Connect your Calendar’ and complete the actions to sync your calendar to Time Off by Deel
- You will see a ‘Success! Your calendar has been connected.’ message and your calendar listed under ‘Your synced calendar’ in Time Off by Deel on Slack
- After your personal calendar is linked, click ‘Add calendar’ under
‘Additional calendars to include?’. The ‘Dedicated Migrations Team Calendar’ calendar ID is
c_fb53e24f590edfb8f313253126123e48d57254dd73266fae6547a2a4890b0c82@group.calendar.google.com
- In Slack, click the
Choosing and recording time off
Please take time off whenever you need it.
A little thoughtfulness on everybody’s part will go a long way toward making it possible for everyone to take their desired days off. As you look to plan your time off, please:
- Ensure you know the company wide paid time off guidelines
- Check the calendar and coordinate with the team to ensure we have sufficient cover for days with low availability.
- schedule your time off as far in advance as you can
- Don’t lock yourself into nonrefundable travel itineraries before you’ve taken steps such as the above to be sure you can get the planned days off
- If possible, plan your time off to avoid disruptions
Communicate through a coverage issue
Please put together a coverage issue for any PTO longer than 5 days. Some Divisions may have other guidance (for example Engineering PTO Coverage). Please follow your Division procedures if there is any conflict, and circulate with the team.
After PTO
See returning from pto.
Resources
- GitLab group
- GitLab project
- Migrating Customers to Dedicated internal documentation
- Cutover Cancellation Impact
adaeb449)
