CtrlProcesses configuration
CtrlProcesses is configured in M-Files Admin, under the vault’s Applications node. Every setting on these pages is generated from the configuration classes in the shipping application.
How the configuration is organised
Section titled “How the configuration is organised”Seven sections. Two of them are the connection to Elements.cloud, one is the viewer a user sees, one is the synchronisation that brings processes in, and one publishes links back. The synchronisation is large enough to have its own heading in the sidebar.
The top level
Section titled “The top level”Sections
Organization ID
The Elements.cloud organization (team) identifier this vault connects to. Required before authentication or synchronization can run.
Authentication
Elements.cloud credentials used to obtain API tokens for the system and end-user contexts.
Process viewer
Settings for the embedded process viewer: tabs, diagram defaults, per-user-group features and object-creation forms.
Person / employee
How the application locates the M-Files object that represents the current user, used when creating objects on their behalf.
Resource
Object type and properties used to read resource details for display in the viewer.
URL publishing
Publishes M-Files object links to Elements.cloud when objects enter the configured workflow states.
Elements synchronization
Imports Elements.cloud releases, process maps, diagrams, activities and related objects into M-Files on a schedule.
Main configuration settings, generated from CtrlProcesses 1.26.9.14.
What each section is for
Section titled “What each section is for”Organization ID and Authentication are the connection. Nothing else can be tested until both work, and both authentication contexts are needed — the system one for the scheduled synchronisation, the end-user one for anything the viewer does on a person’s behalf.
Elements synchronisation is the largest section, and has its own heading in the sidebar. It holds one mapping per Elements.cloud entity, plus the schedules and the pacing settings that apply to all of them.
Process viewer is what a user sees: where the tab appears, which diagram opens for which class of object, what each user group can do, and what is drawn on the activity boxes.
Person / employee and Resource are small sections that tell the application how to find a person and how to read a resource. The person section is required before the viewer can create objects on someone’s behalf.
URL publishing is the outbound direction, driven by workflow states rather than by a schedule.
Logging and email come from the M-Files framework rather than from the application itself, and are identical everywhere, so they are documented once at Logging and email.
Where the schedules live
Section titled “Where the schedules live”Two operations run on schedules, both configured under Elements synchronisation:
| Operation | What it does | How often |
|---|---|---|
| Sync | Imports releases, maps, diagrams, activities and the rest | The main schedule |
| Reconcile | Removes objects whose Elements counterpart is gone | Its own schedule, deliberately less often |
With no triggers configured, neither runs — including at application startup. That is the state to leave them in while mappings are still being built.
The settings most often set wrongly
Section titled “The settings most often set wrongly”Diagram version shown. Master and the latest synced release renumber differently, so a viewer set to master can disagree with the numbers stored on M-Files objects while both are correct. See the note in About.
Minimum request interval. A floor on the gap between API calls, not a delay to add to each one. The synchronisation already detects rate limiting and slows itself down. Leave it at zero unless the tenant has a stated limit.
Maximum retry wait. A budget for the total waiting one run may do. Setting it to zero removes the budget, which means a run can wait indefinitely on a background thread. The setting says so, and it is not a recommendation.
Reference model flags. The list is the opt-in as well as the styling. A reference model with no entry is not flagged on any activity, which is the intended way to show only the standards that matter rather than every one the process is mapped against.