Skip to content

[TEST] test improvements #251

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
paulomach wants to merge 10 commits into
base: 8.0/edge
Choose a base branch
from
Draft

[TEST] test improvements #251

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
9109ef4
testing fix
paulomach Apr 17, 2026
9889ef9
concurrent wait
paulomach Apr 17, 2026
aaff69d
other fixes
paulomach Apr 19, 2026
4e078df
replicate machine improv to k8s
paulomach Apr 20, 2026
8ba7490
test revision for test-app
paulomach Apr 20, 2026
0318b5f
fix app
paulomach Apr 20, 2026
32ee7f9
Merge branch '8.0/edge' of github.com:canonical/mysql-operators into …
paulomach Apr 22, 2026
b93b7de
freeze mysqld instead of kill pod
paulomach Apr 22, 2026
ce99abe
10x default timeout (30s)
paulomach Apr 22, 2026
5bf3ae9
fix calls
paulomach Apr 22, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 9 additions & 5 deletions kubernetes/lib/charms/mysql/v0/mysql.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -152,6 +152,7 @@ def __init__(
MAX_CONNECTIONS_FLOOR = 10
MIM_MEM_BUFFERS = 200 * BYTES_1MiB
ADMIN_PORT = 33062
FORCE_QUORUM_TIMEOUT = 300 # seconds

# Labels are not confidential
SECRET_INTERNAL_LABEL = "secret-id" # noqa: S105
Expand Down Expand Up @@ -2370,13 +2371,16 @@ def force_quorum_from_instance(self) -> None:
instance_def = (
f"{self.cluster_admin_user}:{self.cluster_admin_password}@{self.instance_address}"
)
# TODO: modify/expose timeout in the mysql-shell-client library
address = f"{instance_def}:3306"
command = "\n".join((
f"cluster = dba.get_cluster('{self.cluster_name}')",
f"cluster.force_quorum_using_partition_of('{address}')",
))

executor = self._build_cluster_tcp_executor(self.instance_address)
Comment on lines +2374 to +2381
Copy link
Copy Markdown
Contributor Author

@paulomach paulomach Apr 22, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@sinclert-canonical , testing this here, but how should we expose timeout at this level?

Copy link
Copy Markdown
Contributor

@sinclert-canonical sinclert-canonical Apr 22, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

At the moment, timeouts can only be applied in an operation-per-operation basis, in the BaseExecutor subclasses (see code). That means that timeouts can only be applied when manually executing the Python / SQL operations via an executor class.

I originally thought about exposing this argument one level up the chain, in the client classes, but it feels noisy to do in an operation-per-operation basis (i.e. every class method will have an additional arg), and a bit overkill in their constructors, as not every operation should have the same timeout.

Open to ideas.

try:
self._cluster_client_tcp.force_instance_quorum_into_cluster(
cluster_name=self.cluster_name,
instance_host=instance_def,
instance_port=str(3306),
)
executor.execute_py(command, timeout=FORCE_QUORUM_TIMEOUT)
except ExecutionError as e:
raise MySQLForceQuorumFromInstanceError() from e

Expand Down
21 changes: 11 additions & 10 deletions kubernetes/tests/integration/helpers_ha.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -83,16 +83,17 @@ def check_mysql_units_writes_increment(
app_primary = get_mysql_primary_unit(juju, app_name, app_units[0])
app_max_value = get_mysql_max_written_value(juju, app_name, app_primary)

for unit_name in app_units:
for attempt in Retrying(
reraise=True,
stop=stop_after_delay(5 * MINUTE_SECS),
wait=wait_fixed(10),
):
with attempt:
unit_max_value = get_mysql_max_written_value(juju, app_name, unit_name)
assert unit_max_value > app_max_value, "Writes not incrementing"
app_max_value = unit_max_value
with update_interval(juju, "10s"):
for unit_name in app_units:
for attempt in Retrying(
reraise=True,
stop=stop_after_delay(5 * MINUTE_SECS),
wait=wait_fixed(10),
):
with attempt:
unit_max_value = get_mysql_max_written_value(juju, app_name, unit_name)
assert unit_max_value > app_max_value, "Writes not incrementing"
app_max_value = unit_max_value


def delete_k8s_pod(juju: Juju, unit_name: str) -> None:
Expand Down
2 changes: 1 addition & 1 deletion kubernetes/tests/integration/integration/high_availability/test_async_replication.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -164,7 +164,7 @@ def test_deploy_router_and_app(first_model: str) -> None:
charm=MYSQL_TEST_APP_NAME,
app=MYSQL_TEST_APP_NAME,
base="ubuntu@22.04",
channel="latest/edge",
channel="latest/edge/racing",
num_units=1,
trust=False,
constraints=constraints,
Expand Down
44 changes: 34 additions & 10 deletions kubernetes/tests/integration/integration/high_availability/test_async_replication_upgrade.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@
import os
import time
from collections.abc import Generator
from concurrent.futures import ThreadPoolExecutor
from contextlib import suppress

import jubilant_backports
Expand All @@ -32,6 +33,7 @@
MYSQL_APP_1 = "db1"
MYSQL_APP_2 = "db2"
MYSQL_TEST_APP_NAME = "mysql-test-app"
MYSQL_ROUTER = "mysql-router-k8s"

MINUTE_SECS = 60

Expand Down Expand Up @@ -105,14 +107,23 @@ def test_build_and_deploy(first_model: str, second_model: str, charm: str) -> No
)

logging.info("Waiting for the applications to settle")
model_1.wait(
ready=wait_for_apps_status(jubilant_backports.all_active, MYSQL_APP_1),
timeout=10 * MINUTE_SECS,
)
model_2.wait(
ready=wait_for_apps_status(jubilant_backports.all_active, MYSQL_APP_2),
timeout=10 * MINUTE_SECS,
)

def wait_model_1():
model_1.wait(
ready=wait_for_apps_status(jubilant_backports.all_active, MYSQL_APP_1),
timeout=10 * MINUTE_SECS,
)

def wait_model_2():
model_2.wait(
ready=wait_for_apps_status(jubilant_backports.all_active, MYSQL_APP_2),
timeout=10 * MINUTE_SECS,
)

with ThreadPoolExecutor() as executor:
futures = [executor.submit(wait_model_1), executor.submit(wait_model_2)]
for future in futures:
future.result()
Comment on lines +111 to +126
Copy link
Copy Markdown
Contributor

@astrojuanlu astrojuanlu Apr 23, 2026 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm curious, what does this accomplish? Practically speaking, the execution won't go past this point until both .wait calls return, which is what the sequential code does too.


if path := os.getenv("DATA_SOURCE_PATH"):
logging.info("Loading test database")
Expand Down Expand Up @@ -157,16 +168,29 @@ def test_deploy_test_app(first_model: str) -> None:
charm=MYSQL_TEST_APP_NAME,
app=MYSQL_TEST_APP_NAME,
base="ubuntu@22.04",
channel="latest/edge",
channel="latest/edge/racing",
num_units=1,
constraints=constraints,
)

model_1.deploy(
Copy link
Copy Markdown
Contributor

@astrojuanlu astrojuanlu Apr 23, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Does this intersect with https://warthogs.atlassian.net/browse/DPE-9668 ?

charm=MYSQL_ROUTER,
app=MYSQL_ROUTER,
base="ubuntu@22.04",
channel="8.0/edge",
num_units=1,
trust=True,
)

logging.info("Relating the test application")
model_1.integrate(
f"{MYSQL_APP_1}:database",
f"{MYSQL_ROUTER}:database",
f"{MYSQL_TEST_APP_NAME}:database",
)
model_1.integrate(
f"{MYSQL_ROUTER}:backend-database",
f"{MYSQL_APP_1}:database",
)

model_1.wait(
ready=wait_for_apps_status(jubilant_backports.all_active, MYSQL_TEST_APP_NAME),
Expand Down
92 changes: 58 additions & 34 deletions kubernetes/tests/integration/integration/high_availability/test_primary_switchover.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,6 @@
load_mysql_test_data,
update_interval,
wait_for_apps_status,
wait_for_unit_message,
wait_for_unit_status,
)

Expand All @@ -45,7 +44,7 @@ def test_deploy_highly_available_cluster(juju: Juju, charm: str) -> None:
charm=MYSQL_TEST_APP_NAME,
app=MYSQL_TEST_APP_NAME,
base="ubuntu@22.04",
channel="latest/edge",
channel="latest/edge/racing",
config={"sleep_interval": 300},
num_units=1,
constraints=constraints,
Expand Down Expand Up @@ -111,34 +110,33 @@ def test_cluster_failover_after_majority_loss(juju: Juju) -> None:

non_primary_units = app_units - {primary_unit}

unit_to_promote = non_primary_units.pop()
unit_to_survive = non_primary_units.pop()
units_to_freeze = [non_primary_units.pop(), primary_unit]

logging.info(f"Unit selected for promotion: {unit_to_promote}")
logging.info(f"Unit selected for promotion: {unit_to_survive}")

logging.info("Simulate quorum loss")
units_to_kill = [non_primary_units.pop(), primary_unit]
kill_pods(juju, units_to_kill)
logging.info("Simulating quorum loss via SIGSTOP on mysqld")
freeze_mysql(juju, units_to_freeze)

with update_interval(juju, "45s"):
logging.info("Waiting to settle in error state")
logging.info("Waiting for surviving unit to detect quorum loss")
juju.wait(
ready=lambda status: all((
wait_for_unit_status(app_name, unit_to_promote, "active")(status),
wait_for_unit_message(app_name, units_to_kill[0], "OFFLINE")(status),
wait_for_unit_message(app_name, units_to_kill[1], "OFFLINE")(status),
)),
timeout=15 * MINUTE_SECS,
delay=15,
ready=lambda status: wait_for_unit_status(app_name, unit_to_survive, "active")(status),
timeout=5 * MINUTE_SECS,
delay=10,
)

logging.info("Attempting to promote a unit to primary after quorum loss...")
logging.info("Attempting to promote surviving unit to primary after quorum loss...")
juju.run(
unit=unit_to_promote,
unit=unit_to_survive,
action="promote-to-primary",
params={"scope": "unit", "force": True},
wait=600,
)

logging.info("Resuming frozen units so they can rejoin the cluster")
unfreeze_mysql(juju, units_to_freeze)

with update_interval(juju, "15s"):
logging.info("Waiting for all units to become active after switchover...")
juju.wait(
Expand All @@ -147,20 +145,46 @@ def test_cluster_failover_after_majority_loss(juju: Juju) -> None:
delay=5,
)

assert get_mysql_primary_unit(juju, app_name) == unit_to_promote, "Failover failed"


def kill_pods(juju: Juju, unit_names: list[str]) -> None:
"""Kill the unit pods simultaneously using kubectl."""
pod_names = [get_mysql_instance_label(unit) for unit in unit_names]
cmd = [
"kubectl",
"delete",
"pod",
*pod_names,
"-n",
juju.model,
"--grace-period=0",
"--force",
]
subprocess.run(cmd, check=True)
assert get_mysql_primary_unit(juju, app_name) == unit_to_survive, "Failover failed"


def freeze_mysql(juju: Juju, unit_names: list[str]) -> None:
"""Freeze mysqld in the mysql container via SIGSTOP to simulate unreachable members."""
for unit in unit_names:
pod = get_mysql_instance_label(unit)
subprocess.check_call(
[
"kubectl",
"exec",
pod,
"-n",
juju.model or "testing",
"-c",
"mysql",
"--",
"bash",
"-c",
"kill -STOP $(pgrep -x mysqld)",
],
)


def unfreeze_mysql(juju: Juju, unit_names: list[str]) -> None:
"""Resume frozen mysqld in the mysql container via SIGCONT."""
for unit in unit_names:
pod = get_mysql_instance_label(unit)
subprocess.check_call(
[
"kubectl",
"exec",
pod,
"-n",
juju.model or "testing",
"-c",
"mysql",
"--",
"bash",
"-c",
"kill -CONT $(pgrep -x mysqld)",
],
)
2 changes: 1 addition & 1 deletion ...etes/tests/integration/integration/high_availability/test_replication_data_consistency.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,7 +42,7 @@ def test_deploy_highly_available_cluster(juju: Juju, charm: str) -> None:
charm=MYSQL_TEST_APP_NAME,
app=MYSQL_TEST_APP_NAME,
base="ubuntu@22.04",
channel="latest/edge",
channel="latest/edge/racing",
config={"sleep_interval": 300},
num_units=1,
constraints=constraints,
Expand Down
2 changes: 1 addition & 1 deletion ...rnetes/tests/integration/integration/high_availability/test_replication_data_isolation.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,7 +40,7 @@ def test_deploy_highly_available_cluster(juju: Juju, charm: str) -> None:
charm=MYSQL_TEST_APP_NAME,
app=MYSQL_TEST_APP_NAME,
base="ubuntu@22.04",
channel="latest/edge",
channel="latest/edge/racing",
config={"sleep_interval": 300},
num_units=1,
constraints=constraints,
Expand Down
2 changes: 1 addition & 1 deletion kubernetes/tests/integration/integration/high_availability/test_replication_logs_rotation.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -51,7 +51,7 @@ def test_deploy_highly_available_cluster(juju: Juju, charm: str) -> None:
charm=MYSQL_TEST_APP_NAME,
app=MYSQL_TEST_APP_NAME,
base="ubuntu@22.04",
channel="latest/edge",
channel="latest/edge/racing",
config={"sleep_interval": 300},
num_units=1,
constraints=constraints,
Expand Down
2 changes: 1 addition & 1 deletion kubernetes/tests/integration/integration/high_availability/test_replication_reelection.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -44,7 +44,7 @@ def test_deploy_highly_available_cluster(juju: Juju, charm: str) -> None:
charm=MYSQL_TEST_APP_NAME,
app=MYSQL_TEST_APP_NAME,
base="ubuntu@22.04",
channel="latest/edge",
channel="latest/edge/racing",
config={"sleep_interval": 300},
num_units=1,
constraints=constraints,
Expand Down
2 changes: 1 addition & 1 deletion kubernetes/tests/integration/integration/high_availability/test_replication_scaling.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,7 +42,7 @@ def test_deploy_highly_available_cluster(juju: Juju, charm: str) -> None:
charm=MYSQL_TEST_APP_NAME,
app=MYSQL_TEST_APP_NAME,
base="ubuntu@22.04",
channel="latest/edge",
channel="latest/edge/racing",
config={"sleep_interval": 300},
num_units=1,
constraints=constraints,
Expand Down
4 changes: 2 additions & 2 deletions ...rnetes/tests/integration/integration/high_availability/test_replication_unit_endpoints.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -45,7 +45,7 @@ def test_deploy_highly_available_cluster_1(juju: Juju, charm: str) -> None:
charm="mysql-test-app",
app=MYSQL_TEST_APP_NAME_1,
base="ubuntu@22.04",
channel="latest/edge",
channel="latest/edge/racing",
config={"sleep_interval": 300},
num_units=1,
constraints=constraints,
Expand Down Expand Up @@ -89,7 +89,7 @@ def test_deploy_highly_available_cluster_2(juju: Juju, charm: str) -> None:
charm="mysql-test-app",
app=MYSQL_TEST_APP_NAME_2,
base="ubuntu@22.04",
channel="latest/edge",
channel="latest/edge/racing",
config={"sleep_interval": 300},
num_units=1,
)
Expand Down
2 changes: 1 addition & 1 deletion kubernetes/tests/integration/integration/high_availability/test_self_healing_network_cut.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,7 +48,7 @@ def test_deploy_highly_available_cluster(juju: Juju, charm: str) -> None:
charm=MYSQL_TEST_APP_NAME,
app=MYSQL_TEST_APP_NAME,
base="ubuntu@22.04",
channel="latest/edge",
channel="latest/edge/racing",
config={"sleep_interval": 300},
num_units=1,
constraints=constraints,
Expand Down
2 changes: 1 addition & 1 deletion kubernetes/tests/integration/integration/high_availability/test_self_healing_node_drain.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -46,7 +46,7 @@ def test_deploy_highly_available_cluster(juju: Juju, charm: str) -> None:
charm=MYSQL_TEST_APP_NAME,
app=MYSQL_TEST_APP_NAME,
base="ubuntu@22.04",
channel="latest/edge",
channel="latest/edge/racing",
config={"sleep_interval": 300},
num_units=1,
constraints=constraints,
Expand Down
2 changes: 1 addition & 1 deletion kubernetes/tests/integration/integration/high_availability/test_self_healing_pod.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -44,7 +44,7 @@ def test_deploy_highly_available_cluster(juju: Juju, charm: str) -> None:
charm=MYSQL_TEST_APP_NAME,
app=MYSQL_TEST_APP_NAME,
base="ubuntu@22.04",
channel="latest/edge",
channel="latest/edge/racing",
config={"sleep_interval": 300},
num_units=1,
constraints=constraints,
Expand Down
2 changes: 1 addition & 1 deletion ...netes/tests/integration/integration/high_availability/test_self_healing_process_frozen.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -51,7 +51,7 @@ def test_deploy_highly_available_cluster(juju: Juju, charm: str) -> None:
charm=MYSQL_TEST_APP_NAME,
app=MYSQL_TEST_APP_NAME,
base="ubuntu@22.04",
channel="latest/edge",
channel="latest/edge/racing",
config={"sleep_interval": 300},
num_units=1,
constraints=constraints,
Expand Down
2 changes: 1 addition & 1 deletion ...netes/tests/integration/integration/high_availability/test_self_healing_process_killed.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -50,7 +50,7 @@ def test_deploy_highly_available_cluster(juju: Juju, charm: str) -> None:
charm=MYSQL_TEST_APP_NAME,
app=MYSQL_TEST_APP_NAME,
base="ubuntu@22.04",
channel="latest/edge",
channel="latest/edge/racing",
config={"sleep_interval": 300},
num_units=1,
constraints=constraints,
Expand Down
2 changes: 1 addition & 1 deletion ...tes/tests/integration/integration/high_availability/test_self_healing_restart_graceful.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -46,7 +46,7 @@ def test_deploy_highly_available_cluster(juju: Juju, charm: str) -> None:
charm=MYSQL_TEST_APP_NAME,
app=MYSQL_TEST_APP_NAME,
base="ubuntu@22.04",
channel="latest/edge",
channel="latest/edge/racing",
config={"sleep_interval": 300},
num_units=1,
constraints=constraints,
Expand Down
Loading
Loading