fix: include pod terminated reason/message in claude_truncated error (FAR-95)

Capture the claude container's terminated state (exit code, reason, message,
signal) and surface it in the truncation error so operators see *why* the run
was cut short — e.g. "exit code 137, SIGKILL (commonly OOMKilled),
reason=OOMKilled, message=Memory cgroup out of memory" instead of just a
"truncated" label with no diagnostic context.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
This commit is contained in:
2026-04-26 01:57:43 +00:00
committed by Hugh Commit [agent]
parent 88896eddcf
commit 8474f78fe1
2 changed files with 92 additions and 6 deletions
+32 -2
View File
@@ -60,7 +60,7 @@ vi.mock("@paperclipai/adapter-utils/server-utils", async (importOriginal) => {
});
});
const { isK8s404, buildPartialRunError, classifyOrphan, describePodTerminatedError, streamPodLogsOnce, shouldAbortForCancellation, execute } = await import("./execute.js");
const { isK8s404, buildPartialRunError, classifyOrphan, describePodTerminatedError, describeTruncationCause, streamPodLogsOnce, shouldAbortForCancellation, execute } = await import("./execute.js");
function makeJob(opts: {
runId?: string;
@@ -362,6 +362,33 @@ describe("describePodTerminatedError", () => {
});
});
describe("describeTruncationCause", () => {
it("annotates exit code 137 as SIGKILL/OOM", () => {
const msg = describeTruncationCause({ exitCode: 137, reason: "OOMKilled", message: "Memory cgroup out of memory", signal: null });
expect(msg).toContain("exit code 137");
expect(msg).toContain("SIGKILL");
expect(msg).toContain("OOMKilled");
expect(msg).toContain("Memory cgroup out of memory");
});
it("annotates exit code 143 as SIGTERM", () => {
const msg = describeTruncationCause({ exitCode: 143, reason: null, message: null, signal: null });
expect(msg).toContain("exit code 143");
expect(msg).toContain("SIGTERM");
});
it("falls back to 'pod state unavailable' when state is null", () => {
const msg = describeTruncationCause(null);
expect(msg).toContain("pod state unavailable");
});
it("emits 'no exit code' when exitCode is null but state exists", () => {
const msg = describeTruncationCause({ exitCode: null, reason: "Error", message: null, signal: null });
expect(msg).toContain("no exit code");
expect(msg).toContain("reason=Error");
});
});
describe("execute: all-invalid agent.id (N4)", () => {
it("returns hard error without creating a Job when agent.id sanitizes to null", async () => {
const logs: string[] = [];
@@ -1019,7 +1046,7 @@ describe("execute: happy path", () => {
},
);
mockCoreListPods.mockResolvedValue({
items: [{ metadata: { name: "pod-abc" }, status: { containerStatuses: [{ name: "claude", state: { terminated: { exitCode: 137 } } }] } }],
items: [{ metadata: { name: "pod-abc" }, status: { containerStatuses: [{ name: "claude", state: { terminated: { exitCode: 137, reason: "OOMKilled", message: "Memory cgroup out of memory" } } }] } }],
});
const executePromise = execute(makeCtx());
@@ -1030,6 +1057,9 @@ describe("execute: happy path", () => {
expect(result.errorMessage).toContain("truncated mid-stream");
expect(result.errorMessage).toContain("claude-opus-4-7");
expect(result.errorMessage).toContain("exit code 137");
expect(result.errorMessage).toContain("SIGKILL");
expect(result.errorMessage).toContain("OOMKilled");
expect(result.errorMessage).toContain("Memory cgroup out of memory");
});
it("reconnects log stream and logs status when job completion takes > 3s", async () => {