Beispiel #1
0
def canPing(device):
    """ Check a device is reachable by ping

    Returns False on failure, True on Success
    """
    curRetry = 0
    log.info("INFO: attempting to ping device")
    while curRetry < MAX_RETRIES:
        ret, _ = pingDevice(device)
        if not ret:
            curRetry += 1
            if curRetry == MAX_RETRIES:
                setFlag(
                    errorFile,
                    "Automation Error: Unable to ping device after %s attempts"
                    % MAX_RETRIES)
                return False
            else:
                log.info(
                    "INFO: Unable to ping device after %s try. Sleeping for 90s then retrying"
                    % curRetry)
                time.sleep(90)
        else:
            break  # we're done here
    return True
Beispiel #2
0
def canPing(device):
    """ Check a device is reachable by ping

    Returns False on failure, True on Success
    """
    curRetry = 0
    log.info("INFO: attempting to ping device")
    while curRetry < MAX_RETRIES:
        ret, _ = pingDevice(device)
        if not ret:
            curRetry += 1
            if curRetry == MAX_RETRIES:
                setFlag(errorFile, "Automation Error: Unable to ping device after %s attempts" % MAX_RETRIES)
                return False
            else:
                log.info("INFO: Unable to ping device after %s try. Sleeping for 90s then retrying" % curRetry)
                time.sleep(90)
        else:
            break  # we're done here
    return True
Beispiel #3
0
def checkTegra(master, tegra):
    tegraIP = getIPAddress(tegra)
    tegraPath = os.path.join(options.bbpath, tegra)
    exportFile = os.path.join(tegraPath, '%s_status.log' % tegra)
    errorFile = os.path.join(tegraPath, 'error.flg')
    errorFlag = os.path.isfile(errorFile)
    sTegra = 'OFFLINE'
    sutFound = False
    logTD = None

    status = {'tegra': tegra,
              'active': False,
              'cp': 'OFFLINE',
              'bs': 'OFFLINE',
              'msg': '',
              }

    log.debug('%s: %s' % (tegra, tegraIP))

    if master is None:
        status['environment'] = 's'
        status['master'] = 'localhost'
    else:
        status['environment'] = master['environment'][0]
        status['master'] = 'http://%s:%s' % (
            master['hostname'], master['http_port'])

    fPing, lPing = pingDevice(tegra)
    if fPing:
        try:
            hbSocket = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
            hbSocket.settimeout(float(120))
            hbSocket.connect((tegraIP, 20700))

            sutFound = True

            time.sleep(2)

            hbSocket.send('info\n')

            d = hbSocket.recv(4096)

            log.debug('socket data length %d' % len(d))
            log.debug(d)

            status['active'] = True

            hbSocket.close()
        except:
            status['active'] = False
            dumpException('socket')

        if status['active']:
            sTegra = 'online'
        else:
            sTegra = 'INACTIVE'

        if not sutFound:
            status['msg'] += 'SUTAgent not present;'
    else:
        status['msg'] += '%s %s;' % (lPing[0], lPing[1])

    # Cheat until we have a better check solution for new watch_devices.sh
    status['cp'] = 'active'  # pretend all is well

    if checkSlaveAlive(tegraPath):
        logTD = checkSlaveActive(tegraPath)
        if logTD is not None:
            if (logTD.days > 0) or (logTD.days == 0 and logTD.seconds > 3600):
                status['bs'] = 'INACTIVE'
                status['msg'] += 'BS %dd %ds;' % (logTD.days, logTD.seconds)
            else:
                status['bs'] = 'active'
        else:
            status['bs'] = 'INACTIVE'
    else:
        # scan thru tegra-### dir and see if any buildbot.tac.bug#### files
        # exist but ignore buildbot.tac file itself (except to note that it is
        # missing)
        files = os.listdir(tegraPath)
        found = False
        for f in files:
            if f.startswith('buildbot.tac'):
                found = True
                if len(f) > 12:
                    status['msg'] += '%s;' % f
        if not found:
            status['msg'] += 'buildbot.tac NOT found;'

    if errorFlag:
        status['msg'] += 'error.flg [%s] ' % getLastLine(errorFile)

    s = '%s %s %9s %8s %8s :: %s' % (status['tegra'], status['environment'],
                                     sTegra, status['cp'], status['bs'],
                                     status['msg'])
    ts = time.strftime('%Y-%m-%d %H:%M:%S')
    log.info(s)
    open(exportFile, 'a+').write('%s %s\n' % (ts, s))
    summary(status['tegra'], status['environment'], sTegra, status[
            'cp'], status['bs'], status['msg'], ts, status['master'])

    if errorFlag and options.reset:
        stopProcess(os.path.join(tegraPath, 'twistd.pid'), 'buildslave')

        if not options.reboot:
            try:
                hbSocket = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
                hbSocket.settimeout(float(120))
                hbSocket.connect((tegraIP, 20700))
                hbSocket.send('rebt\n')
                hbSocket.close()
                log.info('rebooting tegra')
            except:
                dumpException('socket')

        if errorFlag:
            log.info('clearing error.flg')
            os.remove(errorFile)

    # Here we try to catch the state where sutagent and cp are inactive that
    # is determined by:
    #     sTegra == 'INACTIVE' and
    #     status['cp'] == 'INACTIVE'
    # status['cp'] will be set to INACTIVE only if logTD.seconds (last time
    # clientproxy updated it's log file) is > 3600

    if options.reboot:
        if not sutFound and status['bs'] != 'active':
            log.info('power cycling tegra')
            reboot_device(tegra)
        else:
            if sTegra == 'OFFLINE' and status['bs'] != 'active':
                log.info('power cycling tegra')
                reboot_device(tegra)

    if options.reset and sTegra == 'INACTIVE' and status['cp'] == 'INACTIVE':
        log.info('stopping hung clientproxy')
        stopDevice(tegra)
        time.sleep(5)
        log.info('starting clientproxy for %s' % tegra)
        os.chdir(tegraPath)
        runCommand(['python', 'clientproxy.py', '-b', '--device=%s' % tegra])
Beispiel #4
0
def checkTegra(master, tegra):
    tegraIP = getIPAddress(tegra)
    tegraPath = os.path.join(options.bbpath, tegra)
    exportFile = os.path.join(tegraPath, '%s_status.log' % tegra)
    errorFile = os.path.join(tegraPath, 'error.flg')
    errorFlag = os.path.isfile(errorFile)
    sTegra = 'OFFLINE'
    sutFound = False
    logTD = None

    status = {
        'tegra': tegra,
        'active': False,
        'cp': 'OFFLINE',
        'bs': 'OFFLINE',
        'msg': '',
    }

    log.debug('%s: %s' % (tegra, tegraIP))

    if master is None:
        status['environment'] = 's'
        status['master'] = 'localhost'
    else:
        status['environment'] = master['environment'][0]
        status['master'] = 'http://%s:%s' % (master['hostname'],
                                             master['http_port'])

    fPing, lPing = pingDevice(tegra)
    if fPing:
        try:
            hbSocket = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
            hbSocket.settimeout(float(120))
            hbSocket.connect((tegraIP, 20700))

            sutFound = True

            time.sleep(2)

            hbSocket.send('info\n')

            d = hbSocket.recv(4096)

            log.debug('socket data length %d' % len(d))
            log.debug(d)

            status['active'] = True

            hbSocket.close()
        except:
            status['active'] = False
            dumpException('socket')

        if status['active']:
            sTegra = 'online'
        else:
            sTegra = 'INACTIVE'

        if not sutFound:
            status['msg'] += 'SUTAgent not present;'
    else:
        status['msg'] += '%s %s;' % (lPing[0], lPing[1])

    # Cheat until we have a better check solution for new watch_devices.sh
    status['cp'] = 'active'  # pretend all is well

    if checkSlaveAlive(tegraPath):
        logTD = checkSlaveActive(tegraPath)
        if logTD is not None:
            if (logTD.days > 0) or (logTD.days == 0 and logTD.seconds > 3600):
                status['bs'] = 'INACTIVE'
                status['msg'] += 'BS %dd %ds;' % (logTD.days, logTD.seconds)
            else:
                status['bs'] = 'active'
        else:
            status['bs'] = 'INACTIVE'
    else:
        # scan thru tegra-### dir and see if any buildbot.tac.bug#### files
        # exist but ignore buildbot.tac file itself (except to note that it is
        # missing)
        files = os.listdir(tegraPath)
        found = False
        for f in files:
            if f.startswith('buildbot.tac'):
                found = True
                if len(f) > 12:
                    status['msg'] += '%s;' % f
        if not found:
            status['msg'] += 'buildbot.tac NOT found;'

    if errorFlag:
        status['msg'] += 'error.flg [%s] ' % getLastLine(errorFile)

    s = '%s %s %9s %8s %8s :: %s' % (status['tegra'], status['environment'],
                                     sTegra, status['cp'], status['bs'],
                                     status['msg'])
    ts = time.strftime('%Y-%m-%d %H:%M:%S')
    log.info(s)
    open(exportFile, 'a+').write('%s %s\n' % (ts, s))
    summary(status['tegra'], status['environment'], sTegra, status['cp'],
            status['bs'], status['msg'], ts, status['master'])

    if errorFlag and options.reset:
        stopProcess(os.path.join(tegraPath, 'twistd.pid'), 'buildslave')

        if not options.reboot:
            try:
                hbSocket = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
                hbSocket.settimeout(float(120))
                hbSocket.connect((tegraIP, 20700))
                hbSocket.send('rebt\n')
                hbSocket.close()
                log.info('rebooting tegra')
            except:
                dumpException('socket')

        if errorFlag:
            log.info('clearing error.flg')
            os.remove(errorFile)

    # Here we try to catch the state where sutagent and cp are inactive that
    # is determined by:
    #     sTegra == 'INACTIVE' and
    #     status['cp'] == 'INACTIVE'
    # status['cp'] will be set to INACTIVE only if logTD.seconds (last time
    # clientproxy updated it's log file) is > 3600

    if options.reboot:
        if not sutFound and status['bs'] != 'active':
            log.info('power cycling tegra')
            reboot_device(tegra)
        else:
            if sTegra == 'OFFLINE' and status['bs'] != 'active':
                log.info('power cycling tegra')
                reboot_device(tegra)

    if options.reset and sTegra == 'INACTIVE' and status['cp'] == 'INACTIVE':
        log.info('stopping hung clientproxy')
        stopDevice(tegra)
        time.sleep(5)
        log.info('starting clientproxy for %s' % tegra)
        os.chdir(tegraPath)
        runCommand(['python', 'clientproxy.py', '-b', '--device=%s' % tegra])