Skip to content

Strategy Executions

A strategy execution is the record of running a remediation strategy on an endpoint. It tracks the full lifecycle from creation through completion, including whether the fix actually worked.

pending → pending_approval → running → succeeded → verified_fixed
→ failed
→ blocked
→ preflight_failed
→ rejected
StatusMeaning
PendingCreated, awaiting dispatch
Pending approvalGovernance requires human approval before running
BlockedGovernance rules blocked execution (blackout window, global pause)
Preflight failedPre-checks failed (disk space, package manager availability)
RunningExecuting on the endpoint via the Furl agent
SucceededExecution completed without errors
FailedExecution encountered an error
RejectedThe end user rejected the approval request

A “succeeded” execution doesn’t necessarily mean the problem is fixed. After execution, Furl verifies whether the finding was actually resolved:

Efficacy statusMeaning
PendingWaiting for the next data sync to check
Verified fixedThe finding is no longer present
Verified not fixedThe finding persists despite “successful” execution
UnknownUnable to verify

This distinction matters: a strategy that runs successfully but doesn’t fix the problem is nearly worthless. A package manager call might exit 0 while the software version doesn’t actually change. Furl detects this and feeds the result back into the strategy’s confidence score — strategies that don’t actually fix things lose confidence over time.

When governance requires approval, the execution enters pending approval and the end user is notified. Approval deadlines are severity-based:

SeverityDeadlineAfter deadline
Critical4 hoursEscalated to admins
High24 hoursEscalated to admins
Medium72 hoursEscalated to admins
Low7 daysEscalated to admins

The end user can approve, reject (with an optional reason), or defer (“remind me later”). If no action is taken before the deadline, the request is escalated.