wally/suits/itest.py

import abc
import time
import logging
import os.path
from typing import Any, List, Optional, Callable, Iterable, cast

from concurrent.futures import ThreadPoolExecutor, wait

from cephlib.node import IRPCNode
from cephlib.units import unit_conversion_coef_f

from ..utils import StopTestError, get_time_interval_printable_info
from ..result_classes import SuiteConfig, JobConfig, TimeSeries, IWallyStorage


logger = logging.getLogger("wally")


__doc__ = "Contains base classes for performance tests"


class PerfTest(metaclass=abc.ABCMeta):
    """Base class for all tests"""
    name = None  # type: str
    max_retry = 3
    retry_time = 30
    job_config_cls = None  # type: type

    def __init__(self, storage: IWallyStorage, suite: SuiteConfig, on_idle: Callable[[], None] = None) -> None:
        self.suite = suite
        self.stop_requested = False
        self.sorted_nodes_ids = sorted(node.node_id for node in self.suite.nodes)
        self.on_idle = on_idle
        self.storage = storage

    def request_stop(self) -> None:
        self.stop_requested = True

    def join_remote(self, path: str) -> str:
        return os.path.join(self.suite.remote_dir, path)

    @abc.abstractmethod
    def run(self) -> None:
        pass

    @abc.abstractmethod
    def format_for_console(self, data: Any) -> str:
        pass


class ThreadedTest(PerfTest, metaclass=abc.ABCMeta):
    """Base class for tests, which spawn separated thread for each node"""

    # max allowed time difference between starts and stops of run of the same test on different test nodes
    # used_max_diff = max((min_run_time * max_rel_time_diff), max_time_diff)
    max_time_diff = 5
    max_rel_time_diff = 0.05
    load_profile_name = None  # type: str

    def __init__(self, *args, **kwargs) -> None:
        PerfTest.__init__(self, *args, **kwargs)
        self.job_configs = None  # type: List[JobConfig]

    @abc.abstractmethod
    def get_expected_runtime(self, iter_cfg: JobConfig) -> Optional[int]:
        pass

    def get_not_done_jobs(self) -> Iterable[JobConfig]:
        jobs_map = {job.storage_id: job for job in self.job_configs}
        already_in_storage = set()
        for db_config in cast(List[JobConfig], self.storage.iter_job(self.suite)):
            if db_config.storage_id in jobs_map:
                job = jobs_map[db_config.storage_id]
                if job != db_config:
                    logger.error("Test info at '%s.%s' is not equal to expected config for iteration %s.%s." +
                                 " Maybe configuration was changed before test was restarted. " +
                                 "DB cfg is:\n    %s\nExpected cfg is:\n    %s\nFix DB or rerun test from beginning",
                                 self.suite.storage_id, job.storage_id, self.name, job.summary,
                                 str(db_config).replace("\n", "\n    "),
                                 str(job).replace("\n", "\n    "))
                    raise StopTestError()

                logger.info("Test iteration %s.%s found in storage and will be skipped", self.name, job.summary)
                already_in_storage.add(db_config.storage_id)

        return [job for job in self.job_configs if job.storage_id not in already_in_storage]

    def run(self) -> None:
        self.storage.put_or_check_suite(self.suite)

        not_in_storage = list(self.get_not_done_jobs())
        if not not_in_storage:
            logger.info("All test iteration in storage already. Skip test")
            return

        logger.debug("Run test %s with profile %r on nodes %s.", self.name,
                                                                 self.load_profile_name,
                                                                 ",".join(self.sorted_nodes_ids))
        logger.debug("Prepare nodes")


        with ThreadPoolExecutor(len(self.suite.nodes)) as pool:
            # config nodes
            list(pool.map(self.config_node, self.suite.nodes))

            run_times = list(map(self.get_expected_runtime, not_in_storage))

            if None not in run_times:
                # +10s - is a rough estimation for additional operations per iteration
                expected_run_time = int(sum(run_times) + 10 * len(not_in_storage))

                exec_time_s, end_dt_s = get_time_interval_printable_info(expected_run_time)
                logger.info("Entire test should takes around %s and finish at %s", exec_time_s, end_dt_s)

            for job in not_in_storage:
                results = []  # type: List[TimeSeries]
                for retry_idx in range(self.max_retry):
                    logger.info("Preparing job %s", job.params.summary)

                    # prepare nodes for new iterations
                    wait([pool.submit(self.prepare_iteration, node, job) for node in self.suite.nodes])

                    expected_job_time = self.get_expected_runtime(job)
                    exec_time_s, end_dt_s = get_time_interval_printable_info(expected_job_time)
                    logger.info("Job should takes around %s and finish at %s", exec_time_s, end_dt_s)

                    jfutures = [pool.submit(self.run_iteration, node, job) for node in self.suite.nodes]
                    failed = False
                    for future in jfutures:
                        try:
                            results.extend(future.result())
                        except EnvironmentError:
                            failed = True

                    if not failed:
                        break

                    if self.max_retry - 1 == retry_idx:
                        logger.exception("Fio failed")
                        raise StopTestError()

                    logger.exception("During fio run")
                    logger.info("Sleeping %ss and retrying job", self.retry_time)
                    time.sleep(self.retry_time)
                    results = []

                # per node jobs start and stop times
                start_times = []  # type: List[int]
                stop_times = []  # type: List[int]

                for ts in results:
                    self.storage.put_ts(ts)
                    if len(ts.times) >= 2:  # type: ignore
                        start_times.append(ts.times[0])
                        stop_times.append(ts.times[-1])

                if len(start_times) > 0:
                    min_start_time = min(start_times)
                    max_start_time = max(start_times)
                    min_stop_time = min(stop_times)

                    max_allowed_time_diff = int((min_stop_time - max_start_time) * self.max_rel_time_diff)
                    max_allowed_time_diff = max(max_allowed_time_diff, self.max_time_diff)

                    if min_start_time + self.max_time_diff < max_allowed_time_diff:
                        logger.warning("Too large difference in %s:%s start time - %s. " +
                                       "Max recommended difference is %s",
                                       self.name, job.summary,
                                       max_start_time - min_start_time, self.max_time_diff)

                    if min_stop_time + self.max_time_diff < max_allowed_time_diff:
                        logger.warning("Too large difference in %s:%s stop time - %s. " +
                                       "Max recommended difference is %s",
                                       self.name, job.summary,
                                       max_start_time - min_start_time, self.max_time_diff)

                    one_s = int(unit_conversion_coef_f('s', results[0].time_units))
                    job.reliable_info_range = (int(max_start_time) + one_s, int(min_stop_time) - one_s)

                self.storage.put_job(self.suite, job)
                self.storage.sync()

                if self.on_idle is not None:
                    self.on_idle()

    @abc.abstractmethod
    def config_node(self, node: IRPCNode) -> None:
        pass

    @abc.abstractmethod
    def prepare_iteration(self, node: IRPCNode, job: JobConfig) -> None:
        pass

    @abc.abstractmethod
    def run_iteration(self, node: IRPCNode, job: JobConfig) -> List[TimeSeries]:
        pass


class TwoScriptTest(ThreadedTest, metaclass=abc.ABCMeta):
    def __init__(self, *dt, **mp) -> None:
        ThreadedTest.__init__(self, *dt, **mp)
        self.prerun_script = self.suite.params['prerun_script']
        self.run_script = self.suite.params['run_script']
        self.prerun_tout = self.suite.params.get('prerun_tout', 3600)
        self.run_tout = self.suite.params.get('run_tout', 3600)
        # TODO: fix job_configs field
        raise NotImplementedError("Fix job configs")

    def get_expected_runtime(self, job: JobConfig) -> Optional[int]:
        return None

    def config_node(self, node: IRPCNode) -> None:
        node.copy_file(self.run_script, self.join_remote(self.run_script))
        node.copy_file(self.prerun_script, self.join_remote(self.prerun_script))

        cmd = self.join_remote(self.prerun_script)
        cmd += ' ' + self.suite.params.get('prerun_opts', '')
        node.run(cmd, timeout=self.prerun_tout)

    def prepare_iteration(self, node: IRPCNode, job: JobConfig) -> None:
        pass

    def run_iteration(self, node: IRPCNode, job: JobConfig) -> List[TimeSeries]:
        # TODO: have to store logs
        cmd = self.join_remote(self.run_script)
        cmd += ' ' + self.suite.params.get('run_opts', '')
        return self.parse_results(node.run(cmd, timeout=self.run_tout))

    @abc.abstractmethod
    def parse_results(self, data: str) -> List[TimeSeries]:
        pass