RITM state briefly changes to Resolved Incomplete before changing to Resolved
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
10-06-2026 08:49 AM
Hi Team,
I am facing an issue with the RITM state (sc_req_item).
The expected behavior is for the RITM to move to Resolved. However, from the Audit History, I can see that the RITM is first being updated to Resolved Incomplete and then to Resolved.
Both updates are happening at the same timestamp and are being performed by System.
Audit history:
- New → Resolved Incomplete
- New → Resolved
- Both updates have the same update number and timestamp.
I have already checked the associated Flow Designer flow and Business Rules, but I have not found any logic explicitly setting the RITM to Resolved Incomplete.
Could you please suggest what else could be updating the RITM state?
Specifically, what would be the best way to identify the exact process/script responsible for setting the RITM to Resolved Incomplete?
Should I check any other areas such as Workflow, Subflows/Actions, Script Includes, legacy workflow activities, or OOB request/RITM state handling?
Any guidance on how to trace the source of this state update would be appreciated.
Thanks!
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
2 weeks ago
Hi,
Your audit history already tells you the most important thing: both entries have the SAME update number and timestamp. That means the state was set twice inside one transaction - first to Resolved Incomplete, and then overwritten to Resolved before the update committed. You are not looking for one process, you are looking for two writers that ran back-to-back in that order, and the final value is simply whichever ran last. That is also why searching for the TEXT "Resolved Incomplete" found nothing: almost nobody sets the label - code sets the underlying state VALUE (Closed Incomplete is value 4 on sc_req_item, Closed Complete is 3; your instance is relabelling those Closed states as Resolved / Resolved Incomplete).
How to identify the exact writers, in order:
1. Reproduce one case in a sub-production instance with debugging aimed at that transaction: close the RITM the same way, with Business Rule debugging / session debug on, and watch the sc_req_item update. The debug log shows each Business Rule that runs on that update in execution order and the values it sets. Two different rules/logic setting state in sequence is exactly the pattern you are hunting.
2. Check the execution records for that specific RITM at that minute, not the definitions: Flow execution details (flow context/logs for that record - each step shows the values the step actually passed to its Update Record action, even if a later step overwrote them), and if a legacy Workflow also runs on sc_req_item, the workflow history for that record (the executed activities, in order). People check the flow and workflow DEFINITIONS and miss that both are executing on the same record.
3. Code Search (Studio > Code Search, or sys code search) for writers you have not covered, searching on the value, not the label: search sc_req_item scripts for state being set to 4 / the Closed Incomplete value, and for script includes that close a request item. Check in particular:
- Task rollup logic on sc_task: logic that closes/updates the parent RITM when a Catalog Task closes. This runs in the same transaction as the task close, as System - your exact pattern. It typically maps "not all tasks Closed Complete" to Closed Incomplete.
- Request-level closure: logic on sc_request that cascades state to its RITMs when the Request completes.
- Any Script Include called by the Flow (custom action/subflow calling a script include will not be visible by reading the Flow alone).
4. Before you change anything, check ALL Catalog Tasks on one affected RITM - including auto-generated tasks and any task that is Closed Skipped or was closed and reopened. The most common underlying story: the task-closure rollup runs first, sees one task that is not Closed Complete (Closed Skipped is the usual suspect, or a task that closes in the same transaction after the check) and correctly - per its rule - writes Closed Incomplete; then your Flow's own closing step runs and overwrites it with Closed Complete, masking it. If that is your case, the Flow is not the bug: there are two components that both believe they own RITM closure, and their mapping of Skipped differs.
Fix direction once you have the two writers from step 1/2: make only one of them set the RITM state (remove the state set from the other), and if the task rollup remains the owner, align its handling of Closed Skipped tasks with the outcome you actually want. Do not just hunt the Incomplete write and delete it - because it happens first and is masked by the second write, the visible symptom would change completely if the Flow's timing/order ever changes.