didericis/bot-bottle
1
0
Fork 0
Code Issues 33 Pull Requests 9 Actions Packages Releases Wiki Activity

feat(reconcile): reprovision CA, git-gate, and egress tokens on gateway bring-up (PRD 0081)
prd-number-check / require-numbered-prds (pull_request) Successful in 10s
Details
tracker-policy-pr / check-pr (pull_request) Successful in 16s
Details
lint / lint (push) Failing after 1m1s
Details
test / unit (pull_request) Successful in 56s
Details
test / image-input-builds (pull_request) Successful in 1m0s
Details
test / integration-docker (pull_request) Failing after 3m7s
Details
test / coverage (pull_request) Has been skipped
Details

Browse Source 
On a gateway cold boot, reconcile every live agent VM against the fresh
gateway: push the new CA into each agent's trust store, re-provision
git-gate repos/creds from the persisted upstreams snapshot, and restore
egress tokens. Per-bottle failures are logged and skipped rather than
aborting the whole reconcile.

New: reconcile.py — attach_bottled_agents_to_gateway, _push_ca,
_reprovision_git_gate, _guest_ip_from_config.

New: git_gate/provision.py writes upstreams.json after key provisioning
so the bring-up reconcile can reconstruct the upstream table without the
manifest.

Wired into FirecrackerInfraService.ensure_running() cold-boot path;
base.py BottleBackend gets a no-op default. Old _reprovision_running_bottles
removed from consolidated_launch.py (superseded by reconcile.py). Tests
migrated and extended.
This commit is contained in:
didericis (claude)
2026-07-28 00:28:26 +00:00
parent 314b30c013
commit d066e4032b
8 changed files with 560 additions and 85 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/unit/test_firecracker_infra.py
+35 -4
View File
@@ -21,6 +21,7 @@ from bot_bottle.backend.firecracker.orchestrator import FirecrackerOrchestrator
# there (not at their home modules).
_ORCH_CLS = "bot_bottle.backend.firecracker.infra.FirecrackerOrchestrator"
_GW_CLS = "bot_bottle.backend.firecracker.infra.FirecrackerGateway"
_RECONCILE = "bot_bottle.backend.firecracker.infra.attach_bottled_agents_to_gateway"
class TestAccessors(unittest.TestCase):
@@ -53,10 +54,13 @@ class TestEnsureRunning(unittest.TestCase):
patch.object(infra_vm, "_health_ok", return_value=True), \
patch.object(infra_vm, "_pidfile_alive", return_value=True), \
patch.object(infra_vm, "stop") as stop, \
patch.object(infra_vm, "ensure_built") as built:
patch.object(infra_vm, "ensure_built") as built, \
patch(_RECONCILE) as reconcile:
url = FirecrackerInfraService().ensure_running()
stop.assert_not_called()
built.assert_not_called()
# Adopt path: no reconcile — gateway state is intact.
reconcile.assert_not_called()
ip = infra_vm.netpool.orch_slot().guest_ip
self.assertEqual(f"http://{ip}:{infra_vm.ORCHESTRATOR_PORT}", url)
@@ -73,7 +77,8 @@ class TestEnsureRunning(unittest.TestCase):
patch.object(infra_vm, "_pidfile_alive", return_value=True), \
patch.object(infra_vm, "stop") as stop, \
patch.object(infra_vm, "ensure_built"), \
patch(_ORCH_CLS) as orch_cls, patch(_GW_CLS) as gw_cls:
patch(_ORCH_CLS) as orch_cls, patch(_GW_CLS) as gw_cls, \
patch(_RECONCILE) as reconcile:
orch_cls.return_value.gateway_url.return_value = "http://10.243.255.1:8099"
orch_cls.return_value.mint_gateway_token.return_value = "gw.jwt"
FirecrackerInfraService().ensure_running()
@@ -85,6 +90,8 @@ class TestEnsureRunning(unittest.TestCase):
"http://10.243.255.1:8099", "gw.jwt")
# The fresh boot records the current version for the next launcher.
self.assertEqual("v-current\n", (d / "booted-version").read_text())
# Cold boot: reconcile fires after gateway is up.
reconcile.assert_called_once()
def test_reboots_when_gateway_dead(self):
# Orchestrator healthy + marker current, but the gateway VM is gone ->
@@ -99,11 +106,13 @@ class TestEnsureRunning(unittest.TestCase):
patch.object(infra_vm, "_pidfile_alive", return_value=False), \
patch.object(infra_vm, "stop") as stop, \
patch.object(infra_vm, "ensure_built"), \
patch(_ORCH_CLS) as orch_cls, patch(_GW_CLS) as gw_cls:
patch(_ORCH_CLS) as orch_cls, patch(_GW_CLS) as gw_cls, \
patch(_RECONCILE) as reconcile:
FirecrackerInfraService().ensure_running()
stop.assert_called_once()
orch_cls.return_value.ensure_running.assert_called_once()
gw_cls.return_value.connect_to_orchestrator.assert_called_once()
reconcile.assert_called_once()
def test_boots_both_when_no_running_pair(self):
with tempfile.TemporaryDirectory() as td:
@@ -112,7 +121,8 @@ class TestEnsureRunning(unittest.TestCase):
patch.object(infra_vm, "_health_ok", return_value=False), \
patch.object(infra_vm, "stop") as stop, \
patch.object(infra_vm, "ensure_built") as built, \
patch(_ORCH_CLS) as orch_cls, patch(_GW_CLS) as gw_cls:
patch(_ORCH_CLS) as orch_cls, patch(_GW_CLS) as gw_cls, \
patch(_RECONCILE) as reconcile:
FirecrackerInfraService().ensure_running()
stop.assert_called_once() # clear stale VMs first
built.assert_called_once()
@@ -120,6 +130,27 @@ class TestEnsureRunning(unittest.TestCase):
# reach the control plane at startup).
orch_cls.return_value.ensure_running.assert_called_once()
gw_cls.return_value.connect_to_orchestrator.assert_called_once()
# Cold boot: reconcile fires to restore CA, git-gate, and egress tokens.
reconcile.assert_called_once()
def test_reconcile_receives_url_and_fresh_gateway(self):
# reconcile gets the orchestrator URL and the gateway service (not the
# class). Verify the args to ensure it can reach the right endpoints.
with tempfile.TemporaryDirectory() as td:
with patch.object(infra_vm, "_infra_dir", return_value=Path(td)), \
patch.object(infra_vm, "expected_version", return_value="v-current"), \
patch.object(infra_vm, "_health_ok", return_value=False), \
patch.object(infra_vm, "stop"), \
patch.object(infra_vm, "ensure_built"), \
patch(_ORCH_CLS) as orch_cls, patch(_GW_CLS) as gw_cls, \
patch(_RECONCILE) as reconcile:
orch_cls.return_value.url.return_value = "http://10.243.255.1:8099"
FirecrackerInfraService().ensure_running()
call_args = reconcile.call_args
# First arg: the orchestrator's host URL.
self.assertEqual("http://10.243.255.1:8099", call_args.args[0])
# Second arg: the gateway service instance (not the class).
self.assertIs(gw_cls.return_value, call_args.args[1])
if __name__ == "__main__":