- Confirm the supported path for the exact source and target components.
- Test real business journeys and integration failures.
- Rehearse restoration, rollback decisions and operational handover.
1. Define the business outcome
Agree why the organization is upgrading. Common goals include improving supportability, replacing aging infrastructure, modernizing the user experience or preparing for a wider platform change. Define how the outcome will be assessed.
“Complete the upgrade” is a milestone. “Meet the agreed search response target under representative load” is an acceptance criterion. Keep the baseline and target measurable, and assign a business or service owner to approve them.
2. Inventory the complete Documentum service
Record every component that the business service depends on, with its version, configuration, owner and connection to the target release.
- Content Server, repositories and client applications.
- Database, operating system and application server components.
- Search, indexing, capture, rendering and transformation services.
- Authentication, single sign-on and relevant identity integrations.
- Workflows, scheduled jobs, custom code and deployment packages.
- Enterprise integrations, including infrequent exports and month-end processing.
Include dependencies outside the Documentum team’s direct control. An interface that runs once a month can still be essential to production acceptance.
3. Confirm a supported upgrade path
Check the applicable OpenText release notes, compatibility information and upgrade instructions for the exact components you run. Establish whether intermediate steps, client changes or prerequisite platform updates are required. Do not assume every source release can move directly to the chosen target.
Review custom applications and integrations early. A supported platform combination does not by itself prove that a custom extension will behave correctly.
4. Build a business-focused test plan
Test complete user journeys: creating content, finding it, opening the correct version, completing an approval and retrieving the required evidence. Use representative business accounts, including accounts that should be denied access.
| Workstream | Decision to resolve | Evidence to review |
|---|---|---|
| Compatibility | Is the component combination and upgrade sequence supported? | Version inventory, release guidance and dependency review |
| Business workflows | Do normal and exception paths behave as required? | Approved end-to-end tests and issue dispositions |
| Integrations | Do success, failure and recovery scenarios work? | Interface tests, error handling and owner approval |
| Performance | Does the target meet agreed usage and batch requirements? | Comparable load tests and service-owner acceptance |
| Recovery | Can the team restore service and handle changed data? | Timed restoration/cutover rehearsal and rollback decision record |
| Operations | Can support teams run the new environment? | Runbooks, monitoring, escalation contacts and handover approval |
Include representative document formats and sizes, peak usage, scheduled jobs and search freshness. A successful administrator login is only an installation check.
5. Rehearse cutover and rollback
Run a timed rehearsal with the people who will perform the production change. Define the activity sequence, checkpoints, communications and decision authority. State when to stop, what triggers rollback and how content or transactions created during the cutover window will be handled.
A backup is only one part of recovery. The team needs a tested restoration procedure, recoverable dependencies and a plan for reconciling changes after the restore point. Confirm the conditions under which rollback remains feasible.
6. Plan support after go-live
Reserve time for post-launch monitoring and issue resolution. Provide architecture and configuration records, known limitations, operating procedures and escalation contacts. Establish monitoring for availability, response times, indexing health and integration processing.
Agree who owns incidents, access requests, scheduled jobs and patching. Close the upgrade when the organization can operate and support the new environment, not simply when the change window ends. If search remains an issue, use the Documentum search diagnostic guide to structure the investigation.
What should an upgrade assessment deliver?
- Current-state findings and an agreed business objective.
- A supported target architecture and upgrade sequence.
- A dependency and risk register with named owners.
- A testing plan, acceptance criteria and approval responsibilities.
- A cutover, restoration and rollback approach.
- Scope assumptions, client dependencies and a phased delivery estimate.
If the project also changes repositories or target platforms, treat the data movement as a separate workstream. Our migration validation guide explains the evidence needed for content, metadata, versions and permissions.
Working checklist
Use this checklist to record your review. Selections stay in this browser view and are not submitted.
0 of 6 checked
Frequently asked questions
Can we upgrade directly from our current Documentum version?
The supported path depends on the source release, target release and connected components. Check the applicable OpenText upgrade instructions and compatibility information before confirming the sequence.
How much downtime does a Documentum upgrade require?
There is no reliable standard outage for every environment. The architecture, data volumes, integrations, upgrade method and rehearsal results determine the change window. Agree it after assessment and a timed rehearsal.
Does a successful installation mean the upgrade is complete?
No. Business workflows, permissions, integrations, performance, recovery and operational readiness still need to meet the agreed acceptance criteria.
What should trigger rollback?
Define triggers before cutover: failed critical tests, unrecoverable interface problems, missed decision checkpoints or failure to meet the agreed service conditions. The decision must also consider changes made since the backup or restore point.
Is a Documentum upgrade the same as a content migration?
Not necessarily. An upgrade changes the software environment; a migration moves or transforms content between environments or platforms. A project may include both, but each needs its own scope, testing and acceptance evidence.



