xref: /aosp_15_r20/external/autotest/client/bin/autotestd_monitor (revision 9c5db1993ded3edbeafc8092d69fe5de2ee02df7)

#!/usr/bin/python3

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import common
import sys, os, signal, time, subprocess, fcntl


def _print_to_file_and_flush(msg, file):
    """Print to the provided file, and flush after printing."""
    print(msg, file=file)
    file.flush()

logdir = sys.argv[1]
stdout_start = int(sys.argv[2])  # number of bytes we can skip on stdout
stderr_start = int(sys.argv[3])  # nubmer of bytes we can skip on stderr
# TODO (crosbug.com/38224)- sbasi: Remove extra logging.
stderr = open(os.path.join(logdir, 'stderr'), 'a', buffering=2)

_print_to_file_and_flush('Entered autotestd_monitor.', file=stderr)

# if any of our tail processes die, the monitor should die too
def kill_self(signum, frame):
    os.kill(os.getpid(), signal.SIGTERM)
signal.signal(signal.SIGCHLD, kill_self)

devnull = open(os.devnull, 'w')

# launch some tail processes to pump the std* streams
def launch_tail(filename, outstream, start):
    path = os.path.join(logdir, filename)
    argv = ['tail', '--retry', '--follow=name', '--bytes=+%d' % start, path]
    # stdout=sys.stdout fails on pre-2.5 python (bug in subprocess module)
    if outstream != subprocess.PIPE and outstream.fileno() == 1:
        return subprocess.Popen(argv, stderr=devnull)
    else:
        return subprocess.Popen(argv, stdout=outstream, stderr=devnull)
stdout_pump = launch_tail('stdout', sys.stdout, stdout_start)
stderr_pump = launch_tail('stderr', sys.stderr, stderr_start)

_print_to_file_and_flush('Finished launching tail subprocesses.', file=stderr)
# wait for logdir/started to exist to be sure autotestd is started
start_time = time.time()
started_file_path = os.path.join(logdir, 'started')
while not os.path.exists(started_file_path):
    time.sleep(1)
    if time.time() - start_time >= 30:
        raise Exception("autotestd failed to start in %s" % logdir)

_print_to_file_and_flush('Finished waiting on autotestd to start.',
                         file=stderr)

# watch the exit code file for an exit
exit_code_file = open(os.path.join(logdir, 'exit_code'))
fcntl.flock(exit_code_file, fcntl.LOCK_EX)
_print_to_file_and_flush('Got lock of exit_code_file.', file=stderr)

try:
    exit_code = exit_code_file.read()
    if len(exit_code) != 4:
        exit_code = -signal.SIGKILL   # autotestd was nuked
    else:
        exit_code = int(exit_code)
finally:
    fcntl.flock(exit_code_file, fcntl.LOCK_UN)
    exit_code_file.close()
    _print_to_file_and_flush('Released lock of exit_code_file and closed it.',
                              file=stderr)

# Give tail a tiny bit of time to finish.
time.sleep(0.01)

_print_to_file_and_flush('Killing child processes.', file=stderr)

# clear the SIGCHLD handler so that killing the tails doesn't kill us
signal.signal(signal.SIGCHLD, signal.SIG_DFL)
os.kill(stdout_pump.pid, signal.SIGTERM)
os.kill(stderr_pump.pid, signal.SIGTERM)

stderr.close()
# exit (with the same code as autotestd)
sys.exit(exit_code)