Store k6 output and add socket info collection

Signed-off-by: Andrey Berezin <a.berezin@yadro.com>
This commit is contained in:
Andrey Berezin 2023-11-20 17:39:15 +03:00
parent ed8f90dfc0
commit 9ab4def44f
9 changed files with 99 additions and 110 deletions

View file

@ -1,4 +1,5 @@
from frostfs_testlib.load.interfaces import Loader, ScenarioRunner from frostfs_testlib.load.interfaces.loader import Loader
from frostfs_testlib.load.interfaces.scenario_runner import ScenarioRunner
from frostfs_testlib.load.load_config import ( from frostfs_testlib.load.load_config import (
EndpointSelectionStrategy, EndpointSelectionStrategy,
K6ProcessAllocationStrategy, K6ProcessAllocationStrategy,

View file

@ -0,0 +1,14 @@
from abc import ABC, abstractmethod
from frostfs_testlib.shell.interfaces import Shell
class Loader(ABC):
@abstractmethod
def get_shell(self) -> Shell:
"""Get shell for the loader"""
@property
@abstractmethod
def ip(self):
"""Get address of the loader"""

View file

@ -1,20 +1,8 @@
from abc import ABC, abstractmethod from abc import ABC, abstractmethod
from frostfs_testlib.load.k6 import K6
from frostfs_testlib.load.load_config import LoadParams from frostfs_testlib.load.load_config import LoadParams
from frostfs_testlib.shell.interfaces import Shell
from frostfs_testlib.storage.cluster import ClusterNode from frostfs_testlib.storage.cluster import ClusterNode
from frostfs_testlib.storage.dataclasses.wallet import WalletInfo
class Loader(ABC):
@abstractmethod
def get_shell(self) -> Shell:
"""Get shell for the loader"""
@property
@abstractmethod
def ip(self):
"""Get address of the loader"""
class ScenarioRunner(ABC): class ScenarioRunner(ABC):
@ -32,6 +20,10 @@ class ScenarioRunner(ABC):
def init_k6_instances(self, load_params: LoadParams, endpoints: list[str], k6_dir: str): def init_k6_instances(self, load_params: LoadParams, endpoints: list[str], k6_dir: str):
"""Init K6 instances""" """Init K6 instances"""
@abstractmethod
def get_k6_instances(self) -> list[K6]:
"""Get K6 instances"""
@abstractmethod @abstractmethod
def start(self): def start(self):
"""Start K6 instances""" """Start K6 instances"""

View file

@ -8,13 +8,8 @@ from time import sleep
from typing import Any from typing import Any
from urllib.parse import urlparse from urllib.parse import urlparse
from frostfs_testlib.load.interfaces import Loader from frostfs_testlib.load.interfaces.loader import Loader
from frostfs_testlib.load.load_config import ( from frostfs_testlib.load.load_config import K6ProcessAllocationStrategy, LoadParams, LoadScenario, LoadType
K6ProcessAllocationStrategy,
LoadParams,
LoadScenario,
LoadType,
)
from frostfs_testlib.processes.remote_process import RemoteProcess from frostfs_testlib.processes.remote_process import RemoteProcess
from frostfs_testlib.reporter import get_reporter from frostfs_testlib.reporter import get_reporter
from frostfs_testlib.resources.common import STORAGE_USER_NAME from frostfs_testlib.resources.common import STORAGE_USER_NAME
@ -59,6 +54,7 @@ class K6:
self.loader: Loader = loader self.loader: Loader = loader
self.shell: Shell = shell self.shell: Shell = shell
self.wallet = wallet self.wallet = wallet
self.preset_output: str = ""
self.summary_json: str = os.path.join( self.summary_json: str = os.path.join(
self.load_params.working_dir, self.load_params.working_dir,
f"{self.load_params.load_id}_{self.load_params.scenario.value}_summary.json", f"{self.load_params.load_id}_{self.load_params.scenario.value}_summary.json",
@ -101,10 +97,10 @@ class K6:
command = " ".join(command_args) command = " ".join(command_args)
result = self.shell.exec(command) result = self.shell.exec(command)
assert ( assert result.return_code == EXIT_RESULT_CODE, f"Return code of preset is not zero: {result.stdout}"
result.return_code == EXIT_RESULT_CODE
), f"Return code of preset is not zero: {result.stdout}" self.preset_output = result.stdout.strip("\n")
return result.stdout.strip("\n") return self.preset_output
@reporter.step_deco("Generate K6 command") @reporter.step_deco("Generate K6 command")
def _generate_env_variables(self) -> str: def _generate_env_variables(self) -> str:
@ -113,31 +109,21 @@ class K6:
env_vars[f"{self.load_params.load_type.value.upper()}_ENDPOINTS"] = ",".join(self.endpoints) env_vars[f"{self.load_params.load_type.value.upper()}_ENDPOINTS"] = ",".join(self.endpoints)
env_vars["SUMMARY_JSON"] = self.summary_json env_vars["SUMMARY_JSON"] = self.summary_json
reporter.attach( reporter.attach("\n".join(f"{param}: {value}" for param, value in env_vars.items()), "K6 ENV variables")
"\n".join(f"{param}: {value}" for param, value in env_vars.items()), "K6 ENV variables" return " ".join([f"-e {param}='{value}'" for param, value in env_vars.items() if value is not None])
)
return " ".join(
[f"-e {param}='{value}'" for param, value in env_vars.items() if value is not None]
)
def start(self) -> None: def start(self) -> None:
with reporter.step( with reporter.step(f"Start load from loader {self.loader.ip} on endpoints {self.endpoints}"):
f"Start load from loader {self.loader.ip} on endpoints {self.endpoints}"
):
self._start_time = int(datetime.utcnow().timestamp()) self._start_time = int(datetime.utcnow().timestamp())
command = ( command = (
f"{self._k6_dir}/k6 run {self._generate_env_variables()} " f"{self._k6_dir}/k6 run {self._generate_env_variables()} "
f"{self._k6_dir}/scenarios/{self.load_params.scenario.value}.js" f"{self._k6_dir}/scenarios/{self.load_params.scenario.value}.js"
) )
user = STORAGE_USER_NAME if self.load_params.scenario == LoadScenario.LOCAL else None user = STORAGE_USER_NAME if self.load_params.scenario == LoadScenario.LOCAL else None
self._k6_process = RemoteProcess.create( self._k6_process = RemoteProcess.create(command, self.shell, self.load_params.working_dir, user)
command, self.shell, self.load_params.working_dir, user
)
def wait_until_finished(self, soft_timeout: int = 0) -> None: def wait_until_finished(self, soft_timeout: int = 0) -> None:
with reporter.step( with reporter.step(f"Wait until load is finished from loader {self.loader.ip} on endpoints {self.endpoints}"):
f"Wait until load is finished from loader {self.loader.ip} on endpoints {self.endpoints}"
):
if self.load_params.scenario == LoadScenario.VERIFY: if self.load_params.scenario == LoadScenario.VERIFY:
timeout = self.load_params.verify_time or 0 timeout = self.load_params.verify_time or 0
else: else:
@ -180,9 +166,11 @@ class K6:
while timeout > 0: while timeout > 0:
if not self._k6_process.running(): if not self._k6_process.running():
return return
remaining_time_hours = f"{timeout//3600}h" if timeout//3600 != 0 else "" remaining_time_hours = f"{timeout//3600}h" if timeout // 3600 != 0 else ""
remaining_time_minutes = f"{timeout//60%60}m" if timeout//60%60 != 0 else "" remaining_time_minutes = f"{timeout//60%60}m" if timeout // 60 % 60 != 0 else ""
logger.info(f"K6 is running. Remaining time {remaining_time_hours}{remaining_time_minutes}{timeout%60}s. Next check after {wait_interval} seconds...") logger.info(
f"K6 is running. Remaining time {remaining_time_hours}{remaining_time_minutes}{timeout%60}s. Next check after {wait_interval} seconds..."
)
sleep(wait_interval) sleep(wait_interval)
timeout -= min(timeout, wait_interval) timeout -= min(timeout, wait_interval)
wait_interval = max( wait_interval = max(
@ -198,9 +186,7 @@ class K6:
raise TimeoutError(f"Expected K6 to finish after {original_timeout} sec.") raise TimeoutError(f"Expected K6 to finish after {original_timeout} sec.")
def get_results(self) -> Any: def get_results(self) -> Any:
with reporter.step( with reporter.step(f"Get load results from loader {self.loader.ip} on endpoints {self.endpoints}"):
f"Get load results from loader {self.loader.ip} on endpoints {self.endpoints}"
):
self.__log_output() self.__log_output()
if not self.summary_json: if not self.summary_json:
@ -231,9 +217,7 @@ class K6:
return False return False
@reporter.step_deco("Wait until K6 process end") @reporter.step_deco("Wait until K6 process end")
@wait_for_success( @wait_for_success(K6_STOP_SIGNAL_TIMEOUT, 15, False, False, "Can not stop K6 process within timeout")
K6_STOP_SIGNAL_TIMEOUT, 15, False, False, "Can not stop K6 process within timeout"
)
def _wait_until_process_end(self): def _wait_until_process_end(self):
return self._k6_process.running() return self._k6_process.running()

View file

@ -1,4 +1,4 @@
from frostfs_testlib.load.interfaces import Loader from frostfs_testlib.load.interfaces.loader import Loader
from frostfs_testlib.resources.load_params import ( from frostfs_testlib.resources.load_params import (
LOAD_NODE_SSH_PASSWORD, LOAD_NODE_SSH_PASSWORD,
LOAD_NODE_SSH_PRIVATE_KEY_PASSPHRASE, LOAD_NODE_SSH_PRIVATE_KEY_PASSPHRASE,

View file

@ -10,7 +10,8 @@ from urllib.parse import urlparse
import yaml import yaml
from frostfs_testlib.cli.frostfs_authmate.authmate import FrostfsAuthmate from frostfs_testlib.cli.frostfs_authmate.authmate import FrostfsAuthmate
from frostfs_testlib.load.interfaces import Loader, ScenarioRunner from frostfs_testlib.load.interfaces.loader import Loader
from frostfs_testlib.load.interfaces.scenario_runner import ScenarioRunner
from frostfs_testlib.load.k6 import K6 from frostfs_testlib.load.k6 import K6
from frostfs_testlib.load.load_config import K6ProcessAllocationStrategy, LoadParams, LoadType from frostfs_testlib.load.load_config import K6ProcessAllocationStrategy, LoadParams, LoadType
from frostfs_testlib.load.loaders import NodeLoader, RemoteLoader from frostfs_testlib.load.loaders import NodeLoader, RemoteLoader
@ -50,6 +51,9 @@ class RunnerBase(ScenarioRunner):
return any([future.result() for future in futures]) return any([future.result() for future in futures])
def get_k6_instances(self):
return self.k6_instances
class DefaultRunner(RunnerBase): class DefaultRunner(RunnerBase):
loaders: list[Loader] loaders: list[Loader]
@ -391,6 +395,7 @@ class LocalRunner(RunnerBase):
return results return results
class S3LocalRunner(LocalRunner): class S3LocalRunner(LocalRunner):
endpoints: list[str] endpoints: list[str]
k6_dir: str k6_dir: str
@ -404,7 +409,8 @@ class S3LocalRunner(LocalRunner):
@reporter.step_deco("Resolve containers in preset") @reporter.step_deco("Resolve containers in preset")
def _resolve_containers_in_preset(self, k6_instance: K6): def _resolve_containers_in_preset(self, k6_instance: K6):
k6_instance.shell.exec( k6_instance.shell.exec(
f"sudo {self.k6_dir}/scenarios/preset/resolve_containers_in_preset.py --endpoint {k6_instance.endpoints[0]} --preset_file {k6_instance.load_params.preset.pregen_json}") f"sudo {self.k6_dir}/scenarios/preset/resolve_containers_in_preset.py --endpoint {k6_instance.endpoints[0]} --preset_file {k6_instance.load_params.preset.pregen_json}"
)
@reporter.step_deco("Init k6 instances") @reporter.step_deco("Init k6 instances")
def init_k6_instances(self, load_params: LoadParams, endpoints: list[str], k6_dir: str): def init_k6_instances(self, load_params: LoadParams, endpoints: list[str], k6_dir: str):
@ -426,9 +432,9 @@ class S3LocalRunner(LocalRunner):
# If we chmod /home/<user_name> folder we can no longer ssh to the node # If we chmod /home/<user_name> folder we can no longer ssh to the node
# !! IMPORTANT !! # !! IMPORTANT !!
if ( if (
load_params.working_dir load_params.working_dir
and not load_params.working_dir == f"/home/{LOAD_NODE_SSH_USER}" and not load_params.working_dir == f"/home/{LOAD_NODE_SSH_USER}"
and not load_params.working_dir == f"/home/{LOAD_NODE_SSH_USER}/" and not load_params.working_dir == f"/home/{LOAD_NODE_SSH_USER}/"
): ):
shell.exec(f"sudo chmod -R 777 {load_params.working_dir}") shell.exec(f"sudo chmod -R 777 {load_params.working_dir}")
@ -444,30 +450,25 @@ class S3LocalRunner(LocalRunner):
@run_optionally(optionals.OPTIONAL_BACKGROUND_LOAD_ENABLED) @run_optionally(optionals.OPTIONAL_BACKGROUND_LOAD_ENABLED)
@reporter.step_deco("Preparation steps") @reporter.step_deco("Preparation steps")
def prepare( def prepare(
self, self,
load_params: LoadParams, load_params: LoadParams,
cluster_nodes: list[ClusterNode], cluster_nodes: list[ClusterNode],
nodes_under_load: list[ClusterNode], nodes_under_load: list[ClusterNode],
k6_dir: str, k6_dir: str,
): ):
self.k6_dir = k6_dir self.k6_dir = k6_dir
with reporter.step("Init s3 client on loaders"): with reporter.step("Init s3 client on loaders"):
storage_node = nodes_under_load[0].service(StorageNode) storage_node = nodes_under_load[0].service(StorageNode)
s3_public_keys = [ s3_public_keys = [node.service(S3Gate).get_wallet_public_key() for node in cluster_nodes]
node.service(S3Gate).get_wallet_public_key() for node in cluster_nodes
]
grpc_peer = storage_node.get_rpc_endpoint() grpc_peer = storage_node.get_rpc_endpoint()
parallel(self.prepare_node, nodes_under_load, k6_dir, load_params, s3_public_keys, grpc_peer) parallel(self.prepare_node, nodes_under_load, k6_dir, load_params, s3_public_keys, grpc_peer)
@reporter.step_deco("Prepare node {cluster_node}") @reporter.step_deco("Prepare node {cluster_node}")
def prepare_node(self, def prepare_node(
cluster_node: ClusterNode, self, cluster_node: ClusterNode, k6_dir: str, load_params: LoadParams, s3_public_keys: list[str], grpc_peer: str
k6_dir: str, ):
load_params: LoadParams, LocalRunner.prepare_node(self, cluster_node, k6_dir, load_params)
s3_public_keys: list[str],
grpc_peer: str):
LocalRunner.prepare_node(self,cluster_node, k6_dir, load_params)
self.endpoints = cluster_node.s3_gate.get_all_endpoints() self.endpoints = cluster_node.s3_gate.get_all_endpoints()
shell = cluster_node.host.get_shell() shell = cluster_node.host.get_shell()
@ -497,9 +498,9 @@ class S3LocalRunner(LocalRunner):
wallet_password=self.wallet.password, wallet_password=self.wallet.password,
).stdout ).stdout
aws_access_key_id = str( aws_access_key_id = str(
re.search( re.search(r"access_key_id.*:\s.(?P<aws_access_key_id>\w*)", issue_secret_output).group(
r"access_key_id.*:\s.(?P<aws_access_key_id>\w*)", issue_secret_output "aws_access_key_id"
).group("aws_access_key_id") )
) )
aws_secret_access_key = str( aws_secret_access_key = str(
re.search( re.search(
@ -509,10 +510,8 @@ class S3LocalRunner(LocalRunner):
) )
configure_input = [ configure_input = [
InteractiveInput(prompt_pattern=r"AWS Access Key ID.*", input=aws_access_key_id), InteractiveInput(prompt_pattern=r"AWS Access Key ID.*", input=aws_access_key_id),
InteractiveInput( InteractiveInput(prompt_pattern=r"AWS Secret Access Key.*", input=aws_secret_access_key),
prompt_pattern=r"AWS Secret Access Key.*", input=aws_secret_access_key
),
InteractiveInput(prompt_pattern=r".*", input=""), InteractiveInput(prompt_pattern=r".*", input=""),
InteractiveInput(prompt_pattern=r".*", input=""), InteractiveInput(prompt_pattern=r".*", input=""),
] ]
shell.exec("aws configure", CommandOptions(interactive_inputs=configure_input)) shell.exec("aws configure", CommandOptions(interactive_inputs=configure_input))

View file

@ -2,13 +2,8 @@ import copy
from typing import Optional from typing import Optional
import frostfs_testlib.resources.optionals as optionals import frostfs_testlib.resources.optionals as optionals
from frostfs_testlib.load.interfaces import ScenarioRunner from frostfs_testlib.load.interfaces.scenario_runner import ScenarioRunner
from frostfs_testlib.load.load_config import ( from frostfs_testlib.load.load_config import EndpointSelectionStrategy, LoadParams, LoadScenario, LoadType
EndpointSelectionStrategy,
LoadParams,
LoadScenario,
LoadType,
)
from frostfs_testlib.load.load_report import LoadReport from frostfs_testlib.load.load_report import LoadReport
from frostfs_testlib.load.load_verifiers import LoadVerifier from frostfs_testlib.load.load_verifiers import LoadVerifier
from frostfs_testlib.reporter import get_reporter from frostfs_testlib.reporter import get_reporter
@ -56,9 +51,7 @@ class BackgroundLoadController:
raise RuntimeError("endpoint_selection_strategy should not be None") raise RuntimeError("endpoint_selection_strategy should not be None")
@run_optionally(optionals.OPTIONAL_BACKGROUND_LOAD_ENABLED, []) @run_optionally(optionals.OPTIONAL_BACKGROUND_LOAD_ENABLED, [])
def _get_endpoints( def _get_endpoints(self, load_type: LoadType, endpoint_selection_strategy: EndpointSelectionStrategy):
self, load_type: LoadType, endpoint_selection_strategy: EndpointSelectionStrategy
):
all_endpoints = { all_endpoints = {
LoadType.gRPC: { LoadType.gRPC: {
EndpointSelectionStrategy.ALL: list( EndpointSelectionStrategy.ALL: list(
@ -85,10 +78,7 @@ class BackgroundLoadController:
) )
), ),
EndpointSelectionStrategy.FIRST: list( EndpointSelectionStrategy.FIRST: list(
set( set(node_under_load.service(S3Gate).get_endpoint() for node_under_load in self.nodes_under_load)
node_under_load.service(S3Gate).get_endpoint()
for node_under_load in self.nodes_under_load
)
), ),
}, },
} }
@ -98,12 +88,8 @@ class BackgroundLoadController:
@run_optionally(optionals.OPTIONAL_BACKGROUND_LOAD_ENABLED) @run_optionally(optionals.OPTIONAL_BACKGROUND_LOAD_ENABLED)
@reporter.step_deco("Prepare load instances") @reporter.step_deco("Prepare load instances")
def prepare(self): def prepare(self):
self.endpoints = self._get_endpoints( self.endpoints = self._get_endpoints(self.load_params.load_type, self.load_params.endpoint_selection_strategy)
self.load_params.load_type, self.load_params.endpoint_selection_strategy self.runner.prepare(self.load_params, self.cluster_nodes, self.nodes_under_load, self.k6_dir)
)
self.runner.prepare(
self.load_params, self.cluster_nodes, self.nodes_under_load, self.k6_dir
)
self.runner.init_k6_instances(self.load_params, self.endpoints, self.k6_dir) self.runner.init_k6_instances(self.load_params, self.endpoints, self.k6_dir)
@run_optionally(optionals.OPTIONAL_BACKGROUND_LOAD_ENABLED) @run_optionally(optionals.OPTIONAL_BACKGROUND_LOAD_ENABLED)

View file

@ -109,12 +109,14 @@ class ClusterStateController:
@run_optionally(optionals.OPTIONAL_FAILOVER_ENABLED) @run_optionally(optionals.OPTIONAL_FAILOVER_ENABLED)
@reporter.step_deco("Start host of node {node}") @reporter.step_deco("Start host of node {node}")
def start_node_host(self, node: ClusterNode): def start_node_host(self, node: ClusterNode, tree_healthcheck: bool = True):
with reporter.step(f"Start host {node.host.config.address}"): with reporter.step(f"Start host {node.host.config.address}"):
node.host.start_host() node.host.start_host()
wait_for_host_online(self.shell, node.storage_node) wait_for_host_online(self.shell, node.storage_node)
self.stopped_nodes.remove(node)
wait_for_node_online(node.storage_node) wait_for_node_online(node.storage_node)
self.stopped_nodes.remove(node) if tree_healthcheck:
self.wait_tree_healthcheck()
@run_optionally(optionals.OPTIONAL_FAILOVER_ENABLED) @run_optionally(optionals.OPTIONAL_FAILOVER_ENABLED)
@reporter.step_deco("Start stopped hosts") @reporter.step_deco("Start stopped hosts")
@ -364,7 +366,7 @@ class ClusterStateController:
@run_optionally(optionals.OPTIONAL_FAILOVER_ENABLED) @run_optionally(optionals.OPTIONAL_FAILOVER_ENABLED)
@reporter.step_deco("Hard reboot host {node} via magic SysRq option") @reporter.step_deco("Hard reboot host {node} via magic SysRq option")
def panic_reboot_host(self, node: ClusterNode, wait_for_return: bool = True): def panic_reboot_host(self, node: ClusterNode, wait_for_return: bool = True, tree_healthcheck: bool = True):
shell = node.host.get_shell() shell = node.host.get_shell()
shell.exec('sudo sh -c "echo 1 > /proc/sys/kernel/sysrq"') shell.exec('sudo sh -c "echo 1 > /proc/sys/kernel/sysrq"')
@ -381,6 +383,8 @@ class ClusterStateController:
time.sleep(10) time.sleep(10)
wait_for_host_online(self.shell, node.storage_node) wait_for_host_online(self.shell, node.storage_node)
wait_for_node_online(node.storage_node) wait_for_node_online(node.storage_node)
if tree_healthcheck:
self.wait_tree_healthcheck()
@reporter.step_deco("Down {interface} to {nodes}") @reporter.step_deco("Down {interface} to {nodes}")
def down_interface(self, nodes: list[ClusterNode], interface: str): def down_interface(self, nodes: list[ClusterNode], interface: str):

View file

@ -12,6 +12,7 @@ from frostfs_testlib.steps.node_management import storage_node_healthcheck
from frostfs_testlib.steps.storage_policy import get_nodes_with_object from frostfs_testlib.steps.storage_policy import get_nodes_with_object
from frostfs_testlib.storage.cluster import Cluster, ClusterNode, NodeBase, StorageNode from frostfs_testlib.storage.cluster import Cluster, ClusterNode, NodeBase, StorageNode
from frostfs_testlib.storage.dataclasses.frostfs_services import MorphChain from frostfs_testlib.storage.dataclasses.frostfs_services import MorphChain
from frostfs_testlib.testing.parallel import parallel
from frostfs_testlib.testing.test_control import retry, wait_for_success from frostfs_testlib.testing.test_control import retry, wait_for_success
from frostfs_testlib.utils.datetime_utils import parse_time from frostfs_testlib.utils.datetime_utils import parse_time
@ -26,12 +27,17 @@ def ping_host(shell: Shell, host: Host):
return shell.exec(f"ping {host.config.address} -c 1", options).return_code return shell.exec(f"ping {host.config.address} -c 1", options).return_code
# TODO: Move to ClusterStateController
@reporter.step_deco("Wait for storage nodes returned to cluster") @reporter.step_deco("Wait for storage nodes returned to cluster")
def wait_all_storage_nodes_returned(shell: Shell, cluster: Cluster) -> None: def wait_all_storage_nodes_returned(shell: Shell, cluster: Cluster) -> None:
for node in cluster.services(StorageNode): nodes = cluster.services(StorageNode)
with reporter.step(f"Run health check for storage at '{node}'"): parallel(_wait_for_storage_node, nodes, shell=shell)
wait_for_host_online(shell, node)
wait_for_node_online(node)
@reporter.step_deco("Run health check for storage at '{node}'")
def _wait_for_storage_node(node: StorageNode, shell: Shell) -> None:
wait_for_host_online(shell, node)
wait_for_node_online(node)
@retry(max_attempts=60, sleep_interval=5, expected_result=0) @retry(max_attempts=60, sleep_interval=5, expected_result=0)
@ -64,10 +70,17 @@ def wait_for_node_online(node: StorageNode):
except Exception as err: except Exception as err:
logger.warning(f"Node healthcheck fails with error {err}") logger.warning(f"Node healthcheck fails with error {err}")
return False return False
finally:
gather_socket_info(node)
return health_check.health_status == "READY" and health_check.network_status == "ONLINE" return health_check.health_status == "READY" and health_check.network_status == "ONLINE"
@reporter.step_deco("Gather socket info for {node}")
def gather_socket_info(node: StorageNode):
node.host.get_shell().exec("ss -tuln | grep 8080", CommandOptions(check=False))
@reporter.step_deco("Check and return status of given service") @reporter.step_deco("Check and return status of given service")
def service_status(service: str, shell: Shell) -> str: def service_status(service: str, shell: Shell) -> str:
return shell.exec(f"sudo systemctl is-active {service}").stdout.rstrip() return shell.exec(f"sudo systemctl is-active {service}").stdout.rstrip()
@ -139,9 +152,7 @@ def multiple_restart(
service_name = node.service(service_type).name service_name = node.service(service_type).name
for _ in range(count): for _ in range(count):
node.host.restart_service(service_name) node.host.restart_service(service_name)
logger.info( logger.info(f"Restart {service_systemctl_name}; sleep {sleep_interval} seconds and continue")
f"Restart {service_systemctl_name}; sleep {sleep_interval} seconds and continue"
)
sleep(sleep_interval) sleep(sleep_interval)
@ -164,9 +175,7 @@ def check_services_status(service_list: list[str], expected_status: str, shell:
@reporter.step_deco("Wait for active status of passed service") @reporter.step_deco("Wait for active status of passed service")
@wait_for_success(60, 5) @wait_for_success(60, 5)
def wait_service_in_desired_state( def wait_service_in_desired_state(service: str, shell: Shell, expected_status: Optional[str] = "active"):
service: str, shell: Shell, expected_status: Optional[str] = "active"
):
real_status = service_status(service=service, shell=shell) real_status = service_status(service=service, shell=shell)
assert ( assert (
expected_status == real_status expected_status == real_status