Decommissioning
The Decommissioning module finds assets that every integration which ever found them has stopped seeing, explains why, and helps you clean them up: tag them as decommissioned and keep their history, take the tag off again, or delete them.
Open it from the sidebar: Modules → Decommissioning. The screen is titled Decommissioned assets.
An asset is called decommissioned only on positive evidence: every integration that ever found it has completed enough successful runs within the threshold, and none of them has seen it since. Anything less is reported with the reason, but never called decommissioned.
Prerequisites
The module works from the integrations in each asset's Detected By field and their run history. For an integration to count as evidence it must:
- discover assets. Enrichment-only integrations, which add data to assets others found, are ignored;
- record an activity timestamp for the asset, such as a last check-in or last logon. Which fields count is part of the field definitions the integration ships; an integration without such a field can never confirm that an asset is gone;
- have run successfully in scans. Only a run that finished as completed counts: that is the only kind of run whose results reach the inventory.
Mark and restore also need the tenant's Tags field, because the
decommissioned mark is the tag decommissioned.
How an asset is judged
Each integration that found the asset gives a verdict:
| Verdict | Meaning |
|---|---|
| Seen | Its newest activity timestamp for the asset is within the threshold. This holds however few runs it has. |
| Too few runs | Fewer successful runs within the threshold than the settings require, so its silence does not count yet. |
| No activity field | The integration reports no activity timestamp at all. |
| No timestamp | It holds no activity timestamp for this asset. |
| Not run recently | Its last successful run came less than the threshold after the asset's last activity, so its silence covers less than the threshold. |
| Not seen | It ran often enough, kept looking for the whole threshold, and never saw the asset. The only evidence that an asset is gone. |
| Excluded | Excluded in the settings. Not counted. |
| Deleted | The integration has been deleted. Not counted. |
| Enrichment only | Not a discovery source. Not counted. |
Where an integration has several activity fields, the newest value counts.
The verdicts give the asset a status:
| Status | When |
|---|---|
| Decommissioned | Every integration that counts says Not seen. |
| Awaiting evidence | Some say Not seen, but at least one other cannot tell yet. |
| Insufficient data | None has seen it, and none that could have is reliable. |
| Active | At least one integration has seen it within the threshold. |
| Not evaluable | No integration counts: none ever found it, or every one that did is excluded or deleted. |
Each status carries a reason (Seen recently, Not seen by any integration, Too few successful runs, No activity evidence, No discovery source, All integrations excluded), which you can filter by.
Settings
Settings opens Decommissioning settings. A tenant that has never saved settings uses the defaults.
| Setting | Default | Range |
|---|---|---|
| Decommissioning threshold (days) | 90 | 1 to 3,650 |
| Minimum successful runs per integration | 3 | 1 to 1,000 |
| Assets to evaluate | Endpoints | Endpoints, Users, or both |
| Scope filter (optional) | Empty | An asset search query, up to 2,000 characters |
| Excluded integrations | None | Any of the tenant's integration instances |
- The threshold is how long an asset must have gone unseen.
- The minimum is the number of fully completed, successful runs an integration needs within the threshold before its silence counts.
- The scope filter narrows the evaluated assets further, for example
fields.type: "Server". - An excluded integration is left out entirely: neither its sightings nor its silence count. A deleted integration drops out of the list by itself.
Saving does not recalculate. The screen says when the settings have changed since the evaluation it shows; run a new one to apply them.
Running an evaluation
Evaluations run on the server, on request. Click Recalculate (or Run the first evaluation on a tenant that has none). The evaluation is queued, reads every asset in scope and the run history of every integration, and can take a while on a large tenant. The page shows its progress and updates when it completes.
- Only one evaluation per tenant is queued or running at a time.
- The screen always shows the newest completed evaluation. A failed evaluation leaves no partial results and the previous one stays on screen, with the failure explained (for example, a scope filter the asset search rejects).
- An evaluation that stops reporting progress for 15 minutes, typically because the server restarted, is marked as failed.
- Velum Edge keeps the results of the newest evaluation and the totals of the last 30 for the trend.
The screen
| Tab | Contents |
|---|---|
| Overview | Tiles for Assets evaluated, each status and Tagged (marked decommissioned and kept, with how many were seen again since). Charts for Assets by status, Time since last activity (Within 7 days, 8-30 days, 31-90 days, 91-180 days, 181-365 days, Over a year, No activity data), Evidence by integration and the decommissioned count over the last evaluations. Clicking a status or an integration opens the assets filtered by it. |
| Assets | Every evaluated asset with its status and reason. Filter by Status, Reason, Integration, Evidence, Kind, Tag and a search, and sort. Click an asset to see each integration's verdict in words, for example "found the asset, 5 successful runs / 3 required, not seen within 90 days". |
| Audit log | Every clean-up action: when, which action, its outcome, the asset, who did it and the detail. Filter by action. |
Evidence by integration rates each integration as Reliable, Too few runs, No activity field, Excluded or Enrichment only, which tells you which integration is holding assets back.
Clean-up actions
Select assets on the Assets tab (or open one) and choose an action. Every action goes through a confirmation and is recorded in the audit log, asset by asset.
| Action | Effect |
|---|---|
| Mark as decommissioned | Adds the tag decommissioned. The asset stays in the inventory with its history. |
| Remove decommissioned tag | Takes the tag off. Assets without the tag are skipped. |
| Delete permanently… | Deletes the asset from the inventory. You must type DELETE to confirm. |
Before marking or deleting, Velum Edge judges each asset again against the current settings and data, because the stored evaluation may be days old. An asset is skipped, with the reason in the audit log, when:
- it no longer exists;
- it is outside the scope the settings now evaluate;
- it is not part of the latest evaluation, or that evaluation did not call it decommissioned;
- it no longer qualifies (for example, an integration has seen it since).
For an asset that is marked or deleted, the audit log keeps the explanation by value: the policy it was judged by and what each integration said. The entry still explains itself after the asset is gone.
A deleted asset is gone from the inventory. An integration that still lists it may create it again on its next run. To take an asset out of service but keep its history, mark it as decommissioned instead.
Limits
| Limit | Value |
|---|---|
| Assets per action request | 1,000 (the console sends larger selections in batches of 1,000) |
| Rows per results page (API) | 500 |
| Scope filter length | 2,000 characters |
| Evaluations kept for the trend | 30 |
Permissions
| Task | Permission |
|---|---|
| View the module, its settings, results and audit log | Asset Management read |
| Save settings, request an evaluation, mark, restore, delete | Asset Management write |
The module must also be switched on for the tenant; see Modules.