fix(sentry): report a native retry failure only when the retry ends (#16865)

## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work.
> - The server sends a Sentry `run_failure` event when an agent run
reaches a terminal failure.
> - A native run that waits for a same-run retry is stored as `failed`
with `nativePhase: "retryable_failure"`. That state is not terminal.
> - `reportRunFailure` reported that pending state. Runs that later
recovered, for example after a workspace copy-back retry, still sent a
run failure event.
> - These events make one catch-all Sentry group noisy and hide the real
terminal failures in it.
> - This pull request skips the pending retry state and reports once
when the retry ends as a terminal failure.
> - The benefit is that run failure events match runs that really
failed.

## Linked Issues or Issue Description

**What happened?**

Two staging runs had a workspace copy-back failure during settlement
(`Daytona syncOut tar command failed ... No space left on device`). The
executor stored each run as `failed` with `nativePhase:
"retryable_failure"` and `errorCode: native_session_interrupted`. Sentry
received a `run_failure` event for each run. About two minutes later,
the copy-back retry succeeded and both runs finished as `succeeded` with
`workspaceFinalizeStatus: "succeeded"`.

**Expected behavior**

A run that waits for a same-run retry must not send a terminal run
failure event. If the retry ends as a terminal failure, the server must
send one event at that time.

**Steps to reproduce**

1. Set `PAPERCLIP_SENTRY_RUN_FAILURES=unexpected`.
2. Start a native run that fails retryably. For example, make the
workspace copy-back fail once.
3. See the `run_failure` event in Sentry, and then see the run finish as
`succeeded` after the retry.

## What Changed

- `reportRunFailure` returns without a report for a native run that is
`failed` with `nativePhase: "retryable_failure"`.
- New `isNewTerminalRunFailure(before, after)`: a status change counts
as before. A failed run that leaves the pending retry state but stays
`failed` also counts.
- Three terminal writers can keep the status `failed` → `failed`. They
now use `isNewTerminalRunFailure`, so they still report once:
  - the resume reconciler's ambiguous-state write;
  - the restart recovery `blocked` write;
  - the committed-result projection (`projectCommittedRun`).
- Result-less retries already reset the run to `running` before they
start, so their terminal write still reports through the existing
status-change guard.
- `doc/observability.md` describes the new rule.

## Verification

- `pnpm exec vitest run
src/services/__tests__/run-failure-report.test.ts
src/__tests__/run-failure-sentry.test.ts
src/__tests__/native-session-resumption.test.ts
src/services/native-runtime/native-run-finalizer-telemetry.test.ts
src/services/native-runtime/native-runner-restart-recovery.integration.test.ts
src/__tests__/heartbeat-status-preserving-telemetry.test.ts
src/services/execution-control-reconciliation.test.ts` (from `server/`):
7 files, 363 passed, 9 skipped.
- New tests:
- `run-failure-report.test.ts`: a pending native retry does not report,
and a `terminal_failure` run does. Unit cases cover
`isNewTerminalRunFailure`.
- `native-session-resumption.test.ts`: a sweep that ends an unreported
pending retry sends one event.
- `native-run-finalizer-telemetry.test.ts`: a committed failure for a
pending-retry row sends one event, and a replay sends none.
- `native-runner-restart-recovery.integration.test.ts`: a restart
recovery `blocked` write for a pending-retry row sends one event.
- I removed each of the three writer changes one at a time. The matching
new test failed each time (`expected [] to have a length of 1` / `to
deeply equal [ [ ObjectContaining{…} ] ]`).
- I did not run the full server `tsc --noEmit` locally. The local
container has a 4 GiB memory limit, and the process was killed (exit
137). CI runs the typecheck.

## Risks

- A run that stays in the pending retry state and never resumes no
longer sends a run failure event. The run stays visible on the board as
failed with a scheduled retry. A later terminal write still reports it.
- The finalizer's `failed` + `retryable_failure` write for a failed
provider result also stops reporting. The terminal write that ends that
retry reports instead: retry exhaustion in `recordRetryableFailure`
reports unconditionally, and commit goes through `finalizeNativeRun` or
`projectCommittedRun`.
- A run that was reported in the pending state before this change can
report once more when it becomes terminal after deploy. This affects
only runs in flight during the upgrade.
- There are no schema, migration, or API changes. Run status, retries,
and recovery do not change.

> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.

## Model Used

- Anthropic Claude Opus 5.5 (`claude-opus-5-5`) in Claude Code, with
tool use (shell, code search, test runs) and a read-only code
exploration subagent. Context window: the standard window for that
model.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-authored-by: devinfoley <139239+devinfoley@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>

doc/observability.md

@@ -715,7 +715,9 @@ backend Sentry DSN and the audited SDK. Set `PAPERCLIP_SENTRY_RUN_FAILURES` to:
 
 Missing, blank, or unrecognized values keep run reporting off. Expected operator
 stops, shutdown interruptions, successful runs, and queued work do not report in
-either enabled mode. The mode does not change run status, retries, recovery,
+either enabled mode. A native run that is `failed` while it waits for a same-run
+retry (`nativePhase` `retryable_failure`) does not report either. It reports once
+if a later write ends the retry as a terminal failure. The mode does not change run status, retries, recovery,
 Telemetry, or OpenTelemetry export. It is process-wide, so an operator can enable
 it on one staging instance without enabling it on other deployments. Restart the
 server after changing deployment environment variables.

server/src/__tests__/native-session-resumption.test.ts

@@ -622,6 +622,46 @@ describe("P6-25 pre-result native session recovery", () => {
 
     expect(mockCaptureRunFailure.mock.calls.slice(captureCallsBefore)).toHaveLength(0);
   });
+
+  it("sends one Sentry event when a sweep ends an unreported pending retry", async () => {
+    const freshRunId = "79000000-0000-4000-8000-000000000103";
+    await db.insert(issues).values({ id: freshRunId, companyId, title: "Independent failure report",
+      status: "in_progress", assigneeAgentId: agentId });
+    // The executor stores a scheduled same-run retry as failed + retryable_failure,
+    // and reportRunFailure skips that write. The sweep's terminal write reports it.
+    await db.insert(heartbeatRuns).values({
+      id: freshRunId,
+      companyId,
+      agentId,
+      nativeIssueId: freshRunId,
+      status: "failed",
+      nativePhase: "retryable_failure",
+      runtimeMode: "native",
+      runtimeModeResolvedAt: new Date(),
+      runnerProfileJson: {
+        nativeExecutionInput: {
+          ...persistedProfile.nativeExecutionInput,
+          binding: { runId: freshRunId },
+        },
+      },
+      contextSnapshot: { issueId: freshRunId },
+    });
+    await db.insert(nativeRunFinalizations).values({
+      runId: freshRunId,
+      companyId,
+      issueId: freshRunId,
+      phase: "retryable_failure",
+      attempt: 1,
+    });
+    const captureCallsBefore = mockCaptureRunFailure.mock.calls.length;
+
+    await claimNativeSessionResumptions({ db, runnerInstanceId: "reaper", runIds: [freshRunId] });
+    await waitForPendingRunFailureReports();
+
+    const newCaptures = mockCaptureRunFailure.mock.calls.slice(captureCallsBefore);
+    expect(newCaptures).toHaveLength(1);
+    expect(newCaptures[0]?.[0]).toMatchObject({ runId: freshRunId, runStatus: "failed" });
+  });
 });
 
 describe.each(["unchanged", "newer_active", "stale_idle"] as const)(

server/src/services/__tests__/run-failure-report.test.ts

@@ -29,7 +29,7 @@ vi.mock("../../log-redaction.js", async (importOriginal) => {
   return { ...actual, redactCurrentUserText: mockRedactCurrentUserText };
 });
 
-import { reportRunFailure, waitForPendingRunFailureReports } from "../run-failure-report.js";
+import { isNewTerminalRunFailure, reportRunFailure, waitForPendingRunFailureReports } from "../run-failure-report.js";
 import { controllerLeaseLossCancellation } from "../run-cancellation.js";
 import { REDACTED_EVENT_VALUE } from "../../redaction.js";
 
@@ -981,6 +981,26 @@ describeEmbeddedPostgres("reportRunFailure", () => {
       .toBe(`the provider process exited with code ${REDACTED_EVENT_VALUE}`);
   });
 
+  it("waits for the terminal write when a native run is failed pending a same-run retry", async () => {
+    await seedCompanyAndAgent();
+    await reportRunFailure(db, buildRun({ runtimeMode: "native", nativePhase: "retryable_failure" }));
+    expect(mockCaptureRunFailure).not.toHaveBeenCalled();
+
+    await reportRunFailure(db, buildRun({ runtimeMode: "native", nativePhase: "terminal_failure" }));
+    expect(mockCaptureRunFailure).toHaveBeenCalledOnce();
+  });
+
+  it("treats leaving a pending native retry as a new terminal failure", () => {
+    const pending = { status: "failed", runtimeMode: "native", nativePhase: "retryable_failure" } as const;
+    expect(isNewTerminalRunFailure({ ...pending, status: "running", nativePhase: "observed" }, pending)).toBe(true);
+    expect(isNewTerminalRunFailure(pending, { ...pending, nativePhase: "terminal_failure" })).toBe(true);
+    expect(isNewTerminalRunFailure(pending, { ...pending, nativePhase: "committed" })).toBe(true);
+    expect(isNewTerminalRunFailure(pending, pending)).toBe(false);
+    expect(isNewTerminalRunFailure({ ...pending, nativePhase: null }, { ...pending, nativePhase: "terminal_failure" })).toBe(false);
+    expect(isNewTerminalRunFailure({ ...pending, runtimeMode: "legacy" }, { ...pending, runtimeMode: "legacy", nativePhase: null }))
+      .toBe(false);
+  });
+
   it("captures nothing for succeeded, cancelled, and interrupted", async () => {
     await seedCompanyAndAgent();
     for (const status of ["succeeded", "cancelled", "interrupted"] as const) {

server/src/services/native-runtime/native-finalization-reconciler.ts

@@ -35,7 +35,7 @@ import {
 import { issueRecoveryActionService } from "../issue-recovery-actions.js";
 import { issueService } from "../issues.js";
 import { emitAgentTaskRun } from "../agent-task-run-telemetry.js";
-import { reportRunFailure } from "../run-failure-report.js";
+import { isNewTerminalRunFailure, reportRunFailure } from "../run-failure-report.js";
 import { resumeNativeWorkspaceFinalization } from "./native-workspace-finalizer.js";
 import { dismissObsoleteNativePolicyReviews } from "./obsolete-policy-reviews.js";
 import {
@@ -519,9 +519,10 @@ export async function claimNativeSessionResumptions(input: {
         }).where(eq(heartbeatRuns.id, row.run.id)).returning();
         // Only a genuine transition into "failed" is a new terminal failure.
         // A candidate that is already "failed" (the filter above admits
-        // both "running" and "failed") must not send a second Sentry event.
+        // both "running" and "failed") must not send a second Sentry event,
+        // unless that earlier write was an unreported pending retry.
         terminalRunToEmit =
-          updatedRun && updatedRun.status !== row.run.status ? updatedRun : null;
+          updatedRun && isNewTerminalRunFailure(row.run, updatedRun) ? updatedRun : null;
         await issueService(tx as unknown as Db).update(
           row.coordinator.issueId,
           { status: "blocked" },

server/src/services/native-runtime/native-restart-recovery.ts

@@ -19,7 +19,7 @@ import { readProcessStartedAt } from "../hot-restart.js";
 import { getServerInfoSnapshot } from "../../server-info.js";
 import { redactSensitiveText } from "../../redaction.js";
 import { issueRecoveryActionService } from "../issue-recovery-actions.js";
-import { reportRunFailure } from "../run-failure-report.js";
+import { isNewTerminalRunFailure, reportRunFailure } from "../run-failure-report.js";
 import { isNativeRunnerOwnershipHeld } from "./native-runner-ownership.js";
 import { nativeRestartInterruptedTurnId, type PersistedNativeSession } from "../../vendor/paperclip-runner/index.js";
 
@@ -895,7 +895,7 @@ export async function claimNativeRestartRecoveries(input: {
           })
           .where(eq(heartbeatRuns.id, row.run.id))
           .returning();
-        if (updatedRun && updatedRun.status !== row.run.status) {
+        if (updatedRun && isNewTerminalRunFailure(row.run, updatedRun)) {
           terminalRunToReport = updatedRun;
         }
         await tx

server/src/services/native-runtime/native-run-finalizer-telemetry.test.ts

@@ -58,6 +58,7 @@ import { restoreNativeWorkspaceBestEffort } from "./native-workspace-best-effort
 import { reconcileNativeFinalizations } from "./native-finalization-reconciler.js";
 import { recoverLegacyUnsafeWorkspaceExports } from "./native-workspace-export-recovery.js";
 import { finalizeNativeRun, recordNativeFinalizationFailure } from "./native-run-finalizer.js";
+import { waitForPendingRunFailureReports } from "../run-failure-report.js";
 import { deliverExecutionStatuses } from "../execution-status-delivery.js";
 import { commitNativeStatusDecision } from "./status-decision-committer.js";
 import { NATIVE_STATUS_ARBITER_POLICY_VERSION, type NativeStatusDecision } from "./status-arbiter.js";
@@ -555,6 +556,34 @@ describeEmbeddedPostgres("native run finalizer / status decision committer — a
     expect(replayed).toEqual(repaired);
   });
 
+  it("reports a committed failure once when the run row was an unreported pending retry", async () => {
+    const fixture = await seedNativeRun();
+    await driveToCompleteResult(fixture, {
+      ...CONTROL_PLANE_CONFORMANCE_TERMINAL,
+      turnTerminalState: "failed",
+      runTerminalState: "failed",
+    });
+    const finalize = () => finalizeNativeRun({ db, runId: fixture.runId, workspaceFinalizeStatus: "succeeded", projectRunStatus: true });
+    await finalize();
+    // The executor stores a scheduled same-run retry as failed + retryable_failure,
+    // and reportRunFailure skips that write.
+    await db.update(heartbeatRuns).set({
+      nativePhase: "retryable_failure",
+    }).where(eq(heartbeatRuns.id, fixture.runId));
+    await waitForPendingRunFailureReports();
+    const captureCallsBefore = mockCaptureRunFailure.mock.calls.length;
+
+    await finalize();
+    await waitForPendingRunFailureReports();
+    expect(captureRunFailureCallsFrom(captureCallsBefore)).toEqual([
+      [expect.objectContaining({ runId: fixture.runId, runStatus: "failed" })],
+    ]);
+
+    await finalize();
+    await waitForPendingRunFailureReports();
+    expect(captureRunFailureCallsFrom(captureCallsBefore)).toHaveLength(1);
+  });
+
   it("retires this run's scheduled finalization retry after successful commit", async () => {
     const fixture = await seedNativeRun();
     await driveToCompleteResult(fixture, CONTROL_PLANE_CONFORMANCE_TERMINAL);

server/src/services/native-runtime/native-run-finalizer.ts

@@ -54,7 +54,7 @@ import {
   readNativeBoardResponseWaitSource,
 } from "./native-board-response-wait.js";
 import { emitAgentTaskRun } from "../agent-task-run-telemetry.js";
-import { reportRunFailure } from "../run-failure-report.js";
+import { isNewTerminalRunFailure, reportRunFailure } from "../run-failure-report.js";
 import { resolveExternalChatResponseWaitAuthorization } from "./chat-attachment-reuse.js";
 import {
   authorizeNativeChatReviewPresentation,
@@ -720,10 +720,13 @@ async function projectCommittedRun(input: {
     )
     .returning();
   // Metadata repairs can preserve the terminal status. Only a genuine status
-  // transition should emit another terminal event.
+  // transition should emit another terminal event. A failed run that leaves a
+  // pending retry still needs its first failure report.
   if (updatedRun && updatedRun.status !== input.run.status) {
     await emitAgentTaskRun(input.db, updatedRun);
     void reportRunFailure(input.db, updatedRun);
+  } else if (updatedRun && isNewTerminalRunFailure(input.run, updatedRun)) {
+    void reportRunFailure(input.db, updatedRun);
   }
 }
 

server/src/services/native-runtime/native-runner-restart-recovery.integration.test.ts

@@ -50,6 +50,7 @@ import { materializeAsset } from "./runtime-context.js";
 import { createRunnerTestTlsTunnel, dockerCommandRunner } from "../../__tests__/helpers/docker-command-runner.js";
 import { createIsolatedTestHome } from "../../__tests__/helpers/isolated-test-home.js";
 import { finalizeNativeRun } from "./native-run-finalizer.js";
+import { waitForPendingRunFailureReports } from "../run-failure-report.js";
 import { closeIdleWarmNativeSessionsForRestart, executePaperclipNativeSession, NativeControllerDetachedForRestartError } from "./native-session-executor.js";
 
 const mockCaptureRunFailure = vi.hoisted(() => vi.fn());
@@ -1455,6 +1456,31 @@ describeEmbeddedPostgres("native runner restart recovery with real processes", (
     expect(await fixture.db.select().from(issueRecoveryActions).where(eq(issueRecoveryActions.sourceIssueId, fixture.issueId))).toHaveLength(1);
   });
 
+  it("reports one event when it blocks a failed run that was an unreported pending retry", async () => {
+    const fixture = await seedRun("PENDING-RETRY-BLOCKED");
+    // The executor stores a scheduled same-run retry as failed + retryable_failure,
+    // and reportRunFailure skips that write. The blocked write reports it.
+    await fixture.db.update(heartbeatRuns).set({
+      status: "failed",
+      nativePhase: "retryable_failure",
+      runnerProfileJson: { sessionCheckpoint: {
+        terminal: { runTerminalState: "failed", turnTerminalState: "failed" },
+        providerSessionId: "unusable-provider-session",
+      } },
+    }).where(eq(heartbeatRuns.id, fixture.runId));
+    await fixture.db.update(nativeRunFinalizations).set({ attempt: 3 }).where(eq(nativeRunFinalizations.runId, fixture.runId));
+    const captureCallsBefore = mockCaptureRunFailure.mock.calls.length;
+
+    expect(await claimNativeRestartRecoveries({
+      db: fixture.db, controller: successor, restartKind: "hard", runIds: [fixture.runId],
+    })).toEqual([{ kind: "blocked", runId: fixture.runId, reason: "provider_checkpoint_permanently_failed" }]);
+    await waitForPendingRunFailureReports();
+
+    expect(mockCaptureRunFailure.mock.calls.slice(captureCallsBefore)).toEqual([
+      [expect.objectContaining({ runId: fixture.runId, runStatus: "failed", errorCode: "native_restart_recovery_blocked" })],
+    ]);
+  });
+
   it("classifies every requested recovery candidate without an implicit 100-run cap", async () => {
     const db = createDb(temporary.connectionString);
     const fixtures: Array<Awaited<ReturnType<typeof seedRun>>> = [];

server/src/services/run-failure-report.ts

@@ -38,6 +38,27 @@ function isRunFailureStatus(status: string): status is RunFailureStatus {
   return status === "failed" || status === "timed_out" || status === "cancelled";
 }
 
+type RunFailureTransitionState = Pick<HeartbeatRun, "status" | "runtimeMode" | "nativePhase">;
+
+/**
+ * A native run that waits for a same-run retry is stored as `failed` with
+ * `nativePhase: "retryable_failure"`. That is not terminal: the retry or the
+ * workspace copy-back retry can still finish the run as succeeded.
+ */
+function isPendingNativeRetry(run: RunFailureTransitionState): boolean {
+  return run.status === "failed" && run.runtimeMode === "native" && run.nativePhase === "retryable_failure";
+}
+
+/**
+ * Whether a status write is a new terminal failure to pass to `reportRunFailure`.
+ * A status change counts. A run that leaves a pending native retry but stays
+ * `failed` also counts, because `reportRunFailure` skipped its earlier write.
+ */
+export function isNewTerminalRunFailure(before: RunFailureTransitionState, after: RunFailureTransitionState): boolean {
+  if (after.status !== before.status) return true;
+  return isPendingNativeRetry(before) && !isPendingNativeRetry(after);
+}
+
 function readTaskId(run: HeartbeatRun): string | null {
   if (run.issueId) return run.issueId;
   if (run.nativeIssueId) return run.nativeIssueId;
@@ -238,7 +259,8 @@ async function isRejectedExternalBotWake(db: Db, run: HeartbeatRun): Promise<boo
 
 /**
  * Report a terminal run failure to Sentry. Returns at once for any status
- * other than failures and unexpected started cancellations. Never throws — a Sentry failure or a
+ * other than failures and unexpected started cancellations, and for a native
+ * run that still waits for a same-run retry. Never throws — a Sentry failure or a
  * database read failure must not change the caller's control flow.
  *
  * Call this beside the caller's own terminal-status write, with
@@ -252,6 +274,8 @@ export function reportRunFailure(db: Db, run: HeartbeatRun, options: RunFailureR
   if (mode === "off") return Promise.resolve();
   if (!isRunFailureStatus(run.status)) return Promise.resolve();
   if (run.status === "cancelled" && !isUnexpectedRunCancellation(run)) return Promise.resolve();
+  // The terminal write that ends the retry reports instead (`isNewTerminalRunFailure`).
+  if (isPendingNativeRetry(run)) return Promise.resolve();
   if (mode === "unexpected" && (
     isMissingSecretConfigurationBlocker(run, options) ||
     isAiConnectionSelectionBlocker(run, options) ||