Skip to content

Upgrades Technical guide

WebCenter Content 12c to 14c: plan the out-of-place upgrade

TL;DR

Oracle documents installing the target distributions in a new Oracle home, then upgrading the existing schemas and domain through the applicable upgrade tools. New binaries do not create a separate database or a reversible traffic switch. Plan an isolated rehearsal, component-specific tasks, business acceptance and a consistent recovery point before scheduling production.

Scope and evidence

Scope
Planning a WebCenter Content 12.2.1.4.0 to 14.1.2.0.0 upgrade; exact patch levels, integrations and topology require confirmation.
Evidence basis
Oracle documentation review and engineering planning guidance. This page is not an environment-tested execution runbook.
Sources checked

Who this is for

The team preparing a WebCenter Content 12.2.1.4.0 to 14.1.2.0.0 change. This is a planning guide for constructing and reviewing the procedure. Exact patch levels, platform certifications and installed components determine the execution steps.

Define what out-of-place means

Oracle's domain upgrade guide uses a new Oracle home for target distributions. The upgrade can still modify the existing repository schemas and domain. Keeping the old home on disk therefore does not make the complete source environment available for an immediate rollback.

Use precise language in the plan:

ItemWhat it establishes
New Oracle homeA separate installation location for target binaries
Reconfigured domainA domain updated through the documented upgrade process
Isolated rehearsal environmentA controlled copy with its own data and dependencies
Parallel production replacementA separate architecture requiring synchronization, cutover and recovery decisions

Select the required architecture from the downtime, data-change and recovery requirements. Do not use these terms interchangeably.

Build the procedure from the component inventory

Confirm the source release and the supported database, operating system and JDK combination. Inventory the installed applications, shared schemas, storage providers, custom components, identity services and ERP connections.

The documented domain process covers preparation, target installation, readiness checking, stopping the source environment, schema upgrade, domain reconfiguration, applicable configuration upgrades and post-upgrade work. Component-specific instructions add requirements to that sequence. Treat the current Oracle guide as the execution authority; this summary does not replace it.

Review the existing schemas before deciding whether any additional schema creation is required. Do not create replacement repository schemas as a routine shortcut.

Include the application work

The Content-specific upgrade guidance includes Capture preparation and application-adapter work. For example, Capture batches and queues need the documented pre-upgrade treatment, and ERP adapters have their own installation or configuration requirements.

Build separate acceptance checks for the actual business process:

  • Content retrieval, revisions, search and permissions.
  • Capture intake and the destination of committed documents.
  • WFR extraction, validation and export, using its separately confirmed version and runtime requirements.
  • Invoice coding, account validation and workflow actions.
  • ERP document links, identifiers and authorization.
  • Cluster routing, session behavior and recovery after a node failure.

A successful schema report is evidence about that upgrade step. It does not prove all of these application behaviors.

Make the recovery boundary explicit

Define a consistent source recovery point across database, content storage, domain configuration and external transactions. Identify what is backed up, how it is restored, how long recovery takes and which dependencies must be stopped or reconciled.

After the target accepts new work, returning to an earlier source copy can lose or duplicate changes unless they are accounted for. Agree whether new processing is held until acceptance, how transactions are reconciled and when the rollback option expires. A load-balancer change alone does not reverse a schema upgrade.

Rehearse recovery as well as upgrade. Keep the observed timings and recovered-state tests with the change record.

Decide readiness from evidence

Use a named owner and expected result for every acceptance test. Separate blockers from agreed follow-up work. Record unresolved certification questions, inaccessible integrations and conditions that could not be reproduced in the rehearsal.

The final go/no-go decision should refer to this evidence and the remaining recovery window. Our validation checklist helps structure the application checks; the isolated rehearsal guide explains the preparation boundary.

How ECMWorks does this

We map the connected estate, prepare the version-specific plan, rehearse it with your team and document the acceptance and recovery decisions. Scope includes the business process around Content Server so the handoff explains both the platform change and what users can do afterward.

Questions

Does out-of-place mean a fully independent parallel environment?

No. A new Oracle home is a separate location for target binaries. Independent domains, databases and content storage require their own architecture and migration plan.

Can the old binaries be restarted against upgraded schemas?

Do not assume that is supported. Recovery must restore the compatible source database, domain and storage state required by the approved plan.

Is RCU creation of new schemas always part of the upgrade?

No. Determine the existing schema inventory and follow the target component documentation. Do not recreate existing repository schemas or prescribe new schemas without an applicable requirement.

Give the upgrade a clear scope.

Tell us the starting release and the change you are considering. We can discuss dependencies, a supported route, rehearsal and the checks your team needs for acceptance.