A result set in the advanced reconciliations engine holds the results for one reconciliation type within a single instance. If a reconciliation definition covers three reconciliation types, each instance produces three result sets, one for each type.
What is an instance?
Each time you run a reconciliation definition, LUSID creates an instance of that definition. If you fix data and rerun it, LUSID reruns the same instance rather than creating a new one. An instance is therefore a run of a reconciliation definition, including any reruns.
Each result set carries its own record in LUSID, and contains results. This means you can track review and approval progress for each reconciliation type separately.
Types of status
Status is tracked at three levels, each more granular than the last:
Entity type | Status field name | Status description | |
|---|---|---|---|
Instance |
| A single summary covering every result set in the instance. |
|
Result set |
| One approval status for each reconciliation type within the instance. |
|
Result |
| Each result inside a result set, one for each match, partial match, or break, tracks whether it needs review and whether it has been reviewed. |
|
How does a result set relate to its instance?
Every result set carries a summary of the reconciliation instance it belongs to. The summary includes a value for the instance’s overall status, which covers every reconciliation type in the instance:
Instance Status | Meaning |
|---|---|
Running | The reconciliation instance is still processing. |
Failures | At least one reconciliation type failed to complete. |
Review & Approval | Results are ready and/or being worked through. |
All Approved | Every reconciliation type in the instance has been approved. |
Locked | The instance is locked and cannot be rerun. |
A result set always shows information on the latest run of the instance at the top level, including:
The run number
When it ran
The dates reconciled
Whether the run succeeded
LUSID stores this history so you can always look back at how a result set looked after an earlier run.
Reviewing results and approving result sets
Reviewing and approving results are two separate steps, each tracked with its own status:
Review individual results within a result set.
Approve the whole result set once each result within it has been reviewed.
Every result set can move through the following states:
Result set status | Meaning |
|---|---|
Under review | The result set has not yet been submitted for approval. |
Pending approval | The reviewer has submitted the set for approval. |
Revisions requested | Approval rejected, one or more results must be reviewed again. |
Approved | Every required approval has been given. |
Not Applicable | The run failed, there’s nothing to review or approve. |
You can configure the review requirements for results in the reconciliation definition; read more. Depending on that review configuration, a result set may need one or more named approvals to move to the Approved state.
Conversely, if the review configuration doesn’t require any approvals, the reviewer’s own submission is enough.
LUSID keeps a full history of every submission and every approval decision made on a result set, including who made them and when.
Result counts
A result set summarises its results in three groups:
Open exceptions: Breaks, partial matches, and partial crosses that aren't resolved
Closed exceptions: Exceptions that were accepted, force matched, or cleared on a rerun
Matches: Matches and crosses
See Understanding Advanced Reconciliations results for more information.