fix(ci): add back server:ci-tests to replay tests (#3976)
Some checks failed
SqlStore Integration Tests / test-postgres (3.12) (push) Failing after 0s
SqlStore Integration Tests / test-postgres (3.13) (push) Failing after 1s
Test External Providers Installed via Module / test-external-providers-from-module (venv) (push) Has been skipped
Integration Tests (Replay) / Integration Tests (, , , client=, ) (push) Failing after 3s
Integration Auth Tests / test-matrix (oauth2_token) (push) Failing after 4s
Pre-commit / pre-commit (push) Failing after 4s
Python Package Build Test / build (3.13) (push) Failing after 5s
Test External API and Providers / test-external (venv) (push) Failing after 6s
Vector IO Integration Tests / test-matrix (push) Failing after 7s
Unit Tests / unit-tests (3.13) (push) Failing after 8s
API Conformance Tests / check-schema-compatibility (push) Successful in 15s
Python Package Build Test / build (3.12) (push) Failing after 39s
Unit Tests / unit-tests (3.12) (push) Failing after 40s
UI Tests / ui-tests (22) (push) Successful in 42s

It is useful for local debugging. If both server and docker are failing,
you can just run server locally to debug which is much easier to do.
This commit is contained in:
Ashwin Bharambe 2025-10-30 11:02:59 -07:00 committed by GitHub
parent 5e20938832
commit 77c8bc6fa7
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
4 changed files with 21 additions and 19 deletions

View file

@ -230,6 +230,7 @@ def instantiate_llama_stack_client(session):
force_restart = os.environ.get("LLAMA_STACK_TEST_FORCE_SERVER_RESTART") == "1"
if force_restart:
print(f"Forcing restart of the server on port {port}")
stop_server_on_port(port)
# Check if port is available

View file

@ -10,7 +10,6 @@ import os
import pytest
import llama_stack.core.telemetry.telemetry as telemetry_module
from llama_stack.testing.api_recorder import patch_httpx_for_test_id
from tests.integration.fixtures.common import instantiate_llama_stack_client
from tests.integration.telemetry.collectors import InMemoryTelemetryManager, OtlpHttpTestCollector
@ -21,33 +20,26 @@ def telemetry_test_collector():
stack_mode = os.environ.get("LLAMA_STACK_TEST_STACK_CONFIG_TYPE", "library_client")
if stack_mode == "server":
# In server mode, the collector must be started and the server is already running.
# The integration test script (scripts/integration-tests.sh) should have set
# LLAMA_STACK_TEST_COLLECTOR_PORT and OTEL_EXPORTER_OTLP_ENDPOINT before starting the server.
try:
collector = OtlpHttpTestCollector()
except RuntimeError as exc:
pytest.skip(str(exc))
env_overrides = {
"OTEL_EXPORTER_OTLP_ENDPOINT": collector.endpoint,
"OTEL_EXPORTER_OTLP_PROTOCOL": "http/protobuf",
"OTEL_BSP_SCHEDULE_DELAY": "200",
"OTEL_BSP_EXPORT_TIMEOUT": "2000",
}
previous_env = {key: os.environ.get(key) for key in env_overrides}
for key, value in env_overrides.items():
os.environ[key] = value
telemetry_module._TRACER_PROVIDER = None
# Verify the collector is listening on the expected endpoint
expected_endpoint = os.environ.get("OTEL_EXPORTER_OTLP_ENDPOINT")
if expected_endpoint and collector.endpoint != expected_endpoint:
pytest.skip(
f"Collector endpoint mismatch: expected {expected_endpoint}, got {collector.endpoint}. "
"Server was likely started before collector."
)
try:
yield collector
finally:
collector.shutdown()
for key, prior in previous_env.items():
if prior is None:
os.environ.pop(key, None)
else:
os.environ[key] = prior
else:
manager = InMemoryTelemetryManager()
try: