feat(state): preserve on crash + always snapshot transcript
Extends the preserve-on-capability-block design to also preserve
state on agent crash, and snapshots the transcript on every
teardown so any resume (crash or capability-block) gets a warm
claude session — not a cold start.
- capability_apply: rename _snapshot_transcript → snapshot_transcript
(public; reused below). No behavior change in the capability path.
- cli/start.py: capture bottle.exec_claude's exit code; while the
container is still alive (inside the launch context):
* always snapshot_transcript(identity)
* if exit_code != 0, mark_preserved(identity)
Then the existing _settle_state runs after teardown.
Now the preservation matrix is:
exit 0 (clean) → snapshot + cleanup state
exit ≠0 (crash, Ctrl-C) → snapshot + preserve + show resume hint
capability-block → (already snapshotted/preserved by apply
before teardown; this path is a no-op
because the container is already gone
by the time exec_claude returns)
snapshot_transcript is best-effort — capability-block's earlier
snapshot is not clobbered when the container is already torn down,
and a missing /home/node/.claude is a warn + skip.
Tested behavior: clean exit doesn't preserve, non-zero exit
(including SIGINT/130 and SIGKILL/137) preserves; empty identity
no-ops both helpers.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
This commit is contained in:
@@ -0,0 +1,93 @@
|
||||
"""Unit: cli/start.py session-end state capture (crash preservation).
|
||||
|
||||
The launch-context machinery is covered by integration; this isolates
|
||||
the post-exec_claude decision: snapshot transcript + mark for
|
||||
preservation if non-zero exit, no-op for clean exit."""
|
||||
|
||||
import tempfile
|
||||
import unittest
|
||||
from pathlib import Path
|
||||
|
||||
from claude_bottle import supervise
|
||||
from claude_bottle.backend.docker import bottle_state
|
||||
from claude_bottle.cli import start as start_mod
|
||||
|
||||
|
||||
class _FakeHomeMixin:
|
||||
def _setup_fake_home(self):
|
||||
self._tmp = tempfile.TemporaryDirectory(prefix="cli-start-settle.")
|
||||
self._original = supervise.claude_bottle_root
|
||||
|
||||
def fake_root() -> Path:
|
||||
return Path(self._tmp.name) / ".claude-bottle"
|
||||
|
||||
supervise.claude_bottle_root = fake_root # type: ignore[assignment]
|
||||
|
||||
def _teardown_fake_home(self):
|
||||
supervise.claude_bottle_root = self._original # type: ignore[assignment]
|
||||
self._tmp.cleanup()
|
||||
|
||||
|
||||
class TestCaptureSessionState(_FakeHomeMixin, unittest.TestCase):
|
||||
def setUp(self):
|
||||
self._setup_fake_home()
|
||||
# Stub the docker-dependent snapshot call so this stays a
|
||||
# unit test. apply_capability_change's integration test
|
||||
# covers the real docker cp path.
|
||||
self._snap_calls: list[str] = []
|
||||
self._orig_snap = start_mod.snapshot_transcript
|
||||
start_mod.snapshot_transcript = lambda identity: (
|
||||
self._snap_calls.append(identity)
|
||||
)
|
||||
|
||||
def tearDown(self):
|
||||
start_mod.snapshot_transcript = self._orig_snap
|
||||
self._teardown_fake_home()
|
||||
|
||||
def test_clean_exit_snapshots_but_does_not_mark(self):
|
||||
start_mod._capture_session_state("dev-abc", exit_code=0)
|
||||
self.assertEqual(["dev-abc"], self._snap_calls)
|
||||
self.assertFalse(bottle_state.is_preserved("dev-abc"))
|
||||
|
||||
def test_crash_snapshots_and_marks(self):
|
||||
start_mod._capture_session_state("dev-abc", exit_code=137)
|
||||
self.assertEqual(["dev-abc"], self._snap_calls)
|
||||
self.assertTrue(bottle_state.is_preserved("dev-abc"))
|
||||
|
||||
def test_ctrl_c_treated_as_crash(self):
|
||||
# SIGINT delivers exit 130; the operator may have Ctrl-C'd
|
||||
# because something went wrong, so we preserve.
|
||||
start_mod._capture_session_state("dev-abc", exit_code=130)
|
||||
self.assertTrue(bottle_state.is_preserved("dev-abc"))
|
||||
|
||||
def test_empty_identity_is_noop(self):
|
||||
# Backends without an identity field shouldn't crash this
|
||||
# path (the _identity_from_plan helper falls back to "").
|
||||
start_mod._capture_session_state("", exit_code=137)
|
||||
self.assertEqual([], self._snap_calls)
|
||||
|
||||
|
||||
class TestSettleState(_FakeHomeMixin, unittest.TestCase):
|
||||
def setUp(self):
|
||||
self._setup_fake_home()
|
||||
|
||||
def tearDown(self):
|
||||
self._teardown_fake_home()
|
||||
|
||||
def test_preserved_state_survives(self):
|
||||
bottle_state.write_per_bottle_dockerfile("dev-abc", "FROM x\n")
|
||||
bottle_state.mark_preserved("dev-abc")
|
||||
start_mod._settle_state("dev-abc")
|
||||
self.assertTrue(bottle_state.bottle_state_dir("dev-abc").is_dir())
|
||||
|
||||
def test_unpreserved_state_is_cleaned(self):
|
||||
bottle_state.write_per_bottle_dockerfile("dev-abc", "FROM x\n")
|
||||
start_mod._settle_state("dev-abc")
|
||||
self.assertFalse(bottle_state.bottle_state_dir("dev-abc").exists())
|
||||
|
||||
def test_empty_identity_is_noop(self):
|
||||
start_mod._settle_state("") # should not raise
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
unittest.main()
|
||||
Reference in New Issue
Block a user