def test_cache(self):
"""Test the caching"""
(handle, filename) = tempfile.mkstemp()
os.unlink(filename)
n = SimpleNagios(_cache=filename, _cache_user=self.nagios_user)
message = "mywarning"
n.warning(message)
os.close(handle)
self.buffo.seek(0)
self.buffo.truncate(0)
raised_exception = None
try:
reporter_test = NagiosReporter('test_cache', filename, -1,
self.nagios_user)
reporter_test.report_and_exit()
except SystemExit as err:
raised_exception = err
bo = self.buffo.getvalue().rstrip()
self.assertEqual(bo, "WARNING %s" % message)
self.assertEqual(raised_exception.code, NAGIOS_EXIT_WARNING[0])
statres = os.stat(filename)
self.assertFalse(statres.st_mode & stat.S_IROTH)
def test_cache(self):
"""Test the caching mechanism in the reporter."""
length = random.randint(1, 30)
exit_code = random.randint(0, 3)
threshold = random.randint(0, 10)
message = ''.join(random.choice(string.printable) for x in range(length))
message = message.rstrip()
(handle, filename) = tempfile.mkstemp()
os.unlink(filename)
os.close(handle)
reporter = NagiosReporter('test_cache', filename, threshold, self.nagios_user)
nagios_exit = [NAGIOS_EXIT_OK, NAGIOS_EXIT_WARNING, NAGIOS_EXIT_CRITICAL, NAGIOS_EXIT_UNKNOWN][exit_code]
reporter.cache(nagios_exit, message)
(handle, output_filename) = tempfile.mkstemp()
os.close(handle)
try:
old_stdout = sys.stdout
buffer = StringIO.StringIO()
sys.stdout = buffer
reporter_test = NagiosReporter('test_cache', filename, threshold, self.nagios_user)
reporter_test.report_and_exit()
except SystemExit, err:
line = buffer.getvalue().rstrip()
sys.stdout = old_stdout
buffer.close()
self.assertTrue(err.code == nagios_exit[0])
self.assertTrue(line == "%s %s" % (nagios_exit[1], message))
def test_cache(self, exit_code, message, threshold):
"""Test the caching mechanism in the reporter."""
message = message.rstrip()
if message == '':
return
(handle, filename) = tempfile.mkstemp()
os.unlink(filename)
reporter = NagiosReporter('test_cache', filename, threshold,
self.nagios_user)
nagios_exit = [
NAGIOS_EXIT_OK, NAGIOS_EXIT_WARNING, NAGIOS_EXIT_CRITICAL,
NAGIOS_EXIT_UNKNOWN
][exit_code]
reporter.cache(nagios_exit, message)
(handle, output_filename) = tempfile.mkstemp()
os.close(handle)
try:
old_stdout = sys.stdout
buffer = StringIO.StringIO()
sys.stdout = buffer
reporter_test = NagiosReporter('test_cache', filename, threshold,
self.nagios_user)
reporter_test.report_and_exit()
except SystemExit, err:
line = buffer.getvalue().rstrip()
sys.stdout = old_stdout
buffer.close()
self.assertTrue(err.code == nagios_exit[0])
self.assertTrue(line == "%s %s" % (nagios_exit[1], message))
def test_threshold(self, message="Hello"):
"""Test the threshold borking mechanism in the reporter."""
message = message.rstrip()
threshold = 1
if message == '':
return
(handle, filename) = tempfile.mkstemp()
os.unlink(filename)
reporter = NagiosReporter('test_cache', filename, threshold, self.nagios_user)
# redirect stdout
old_stdout = sys.stdout
buff = StringIO.StringIO()
sys.stdout = buff
nagios_exit = NAGIOS_EXIT_OK
reporter.cache(nagios_exit, message)
os.close(handle)
try:
reporter_test = NagiosReporter('test_cache', filename, threshold, self.nagios_user)
reporter_test.report_and_exit()
except SystemExit, err:
pass
def main(args):
"""Main script."""
options = {
'nagios': ('print out nagion information', None, 'store_true', False, 'n'),
'nagios_check_filename': ('filename of where the nagios check data is stored', str, 'store', NAGIOS_CHECK_FILENAME),
'nagios_check_interval_threshold': ('threshold of nagios checks timing out', None, 'store', NAGIOS_CHECK_INTERVAL_THRESHOLD),
'mail-report': ('mail a report to the hpc-admin list with job list for gracing or inactive users',
None, 'store_true', False),
'ha': ('high-availability master IP address', None, 'store', None),
'dry-run': ('do not make any updates whatsoever', None, 'store_true', False),
}
opts = simple_option(options)
nagios_reporter = NagiosReporter(NAGIOS_HEADER, NAGIOS_CHECK_FILENAME, NAGIOS_CHECK_INTERVAL_THRESHOLD)
if opts.options.nagios:
nagios_reporter.report_and_exit()
sys.exit(0) # not reached
if not proceed_on_ha_service(opts.options.ha):
logger.warning("Not running on the target host in the HA setup. Stopping.")
nagios_reporter(NAGIOS_EXIT_WARNING,
NagiosResult("Not running on the HA master."))
sys.exit(NAGIOS_EXIT_WARNING)
try:
vsc_config = VscConfiguration()
LdapQuery(vsc_config)
grace_users = get_user_with_status('grace')
inactive_users = get_user_with_status('inactive')
pbs_query = PBSQuery()
t = time.ctime()
jobs = pbs_query.getjobs() # we just get them all
removed_queued = remove_queued_jobs(jobs, grace_users, inactive_users, opts.options.dry_run)
removed_running = remove_running_jobs(jobs, inactive_users, opts.options.dry_run)
if opts.options.mail_report and not opts.options.dry_run:
if len(removed_queued) > 0 or len(removed_running) > 0:
mail_report(t, removed_queued, removed_running)
except Exception, err:
logger.exception("Something went wrong: {err}".format(err=err))
nagios_reporter.cache(NAGIOS_EXIT_CRITICAL,
NagiosResult("Script failed, check log file ({logfile})".format(logfile=PBS_CHECK_LOG_FILE)))
sys.exit(NAGIOS_EXIT_CRITICAL)
def test_world_readable(self):
"""Test world readable cache"""
(handle, filename) = tempfile.mkstemp()
os.unlink(filename)
n = SimpleNagios(_cache=filename, _cache_user=self.nagios_user, _world_readable=True)
n.ok("test")
os.close(handle)
try:
reporter_test = NagiosReporter('test_cache', filename, -1, self.nagios_user)
reporter_test.report_and_exit()
except SystemExit:
pass
statres = os.stat(filename)
self.assertTrue(statres.st_mode & stat.S_IROTH)
def test_cache(self):
"""Test the caching"""
(handle, filename) = tempfile.mkstemp()
os.unlink(filename)
n = SimpleNagios(_cache=filename, _cache_user=self.nagios_user)
message = "mywarning"
n.warning(message)
os.close(handle)
self.buffo.seek(0)
self.buffo.truncate(0)
try:
reporter_test = NagiosReporter('test_cache', filename, -1, self.nagios_user)
reporter_test.report_and_exit()
except SystemExit, e:
pass
def test_cache(self):
"""Test the caching"""
(handle, filename) = tempfile.mkstemp()
os.unlink(filename)
n = SimpleNagios(_cache=filename, _cache_user=self.nagios_user)
message = "mywarning"
n.warning(message)
os.close(handle)
self.buffo.seek(0)
self.buffo.truncate(0)
try:
reporter_test = NagiosReporter('test_cache', filename, -1,
self.nagios_user)
reporter_test.report_and_exit()
except SystemExit, e:
pass
def test_world_readable(self):
"""Test world readable cache"""
(handle, filename) = tempfile.mkstemp()
os.unlink(filename)
n = SimpleNagios(_cache=filename,
_cache_user=self.nagios_user,
_world_readable=True)
n.ok("test")
os.close(handle)
try:
reporter_test = NagiosReporter('test_cache', filename, -1,
self.nagios_user)
reporter_test.report_and_exit()
except SystemExit:
pass
statres = os.stat(filename)
self.assertTrue(statres.st_mode & stat.S_IROTH)
def test_cache(self):
"""Test the caching mechanism in the reporter."""
length = random.randint(1, 30)
exit_code = random.randint(0, 3)
threshold = random.randint(0, 10)
message = ''.join(random.choice(string.printable) for x in range(length))
message = message.rstrip()
(handle, filename) = tempfile.mkstemp()
os.unlink(filename)
os.close(handle)
reporter = NagiosReporter('test_cache', filename, threshold, self.nagios_user)
nagios_exit = [NAGIOS_EXIT_OK, NAGIOS_EXIT_WARNING, NAGIOS_EXIT_CRITICAL, NAGIOS_EXIT_UNKNOWN][exit_code]
reporter.cache(nagios_exit, message)
(handle, output_filename) = tempfile.mkstemp()
os.close(handle)
try:
old_stdout = sys.stdout
buffer = StringIO.StringIO()
sys.stdout = buffer
reporter_test = NagiosReporter('test_cache', filename, threshold, self.nagios_user)
reporter_test.report_and_exit()
except SystemExit, err:
line = buffer.getvalue().rstrip()
sys.stdout = old_stdout
buffer.close()
self.assertTrue(err.code == nagios_exit[0])
self.assertTrue(line == "%s %s" % (nagios_exit[1], message))
def test_threshold(self, message="Hello"):
"""Test the threshold borking mechanism in the reporter."""
message = message.rstrip()
threshold = 1
if message == '':
return
(handle, filename) = tempfile.mkstemp()
os.unlink(filename)
reporter = NagiosReporter('test_cache', filename, threshold, self.nagios_user)
# redirect stdout
old_stdout = sys.stdout
buff = StringIO.StringIO()
sys.stdout = buff
nagios_exit = NAGIOS_EXIT_OK
reporter.cache(nagios_exit, message)
os.close(handle)
try:
reporter_test = NagiosReporter('test_cache', filename, threshold, self.nagios_user)
reporter_test.report_and_exit()
except SystemExit, err:
pass
def test_cache(self, exit_code, message, threshold):
"""Test the caching mechanism in the reporter."""
message = message.rstrip()
if message == '':
return
(handle, filename) = tempfile.mkstemp()
os.unlink(filename)
reporter = NagiosReporter('test_cache', filename, threshold, self.nagios_user )
nagios_exit = [NAGIOS_EXIT_OK, NAGIOS_EXIT_WARNING, NAGIOS_EXIT_CRITICAL, NAGIOS_EXIT_UNKNOWN][exit_code]
reporter.cache(nagios_exit, message)
(handle, output_filename) = tempfile.mkstemp()
os.close(handle)
try:
old_stdout = sys.stdout
buffer = StringIO.StringIO()
sys.stdout = buffer
reporter_test = NagiosReporter('test_cache', filename, threshold, self.nagios_user)
reporter_test.report_and_exit()
except SystemExit, err:
line = buffer.getvalue().rstrip()
sys.stdout = old_stdout
buffer.close()
self.assertTrue(err.code == nagios_exit[0])
self.assertTrue(line == "%s %s" % (nagios_exit[1], message))
def test_threshold(self, message="Hello"):
"""Test the threshold borking mechanism in the reporter."""
message = message.rstrip()
threshold = 1
if message == '':
return
(handle, filename) = tempfile.mkstemp()
os.unlink(filename)
reporter = NagiosReporter('test_cache', filename, threshold,
self.nagios_user)
nagios_exit = NAGIOS_EXIT_OK
reporter.cache(nagios_exit, message)
os.close(handle)
try:
reporter_test = NagiosReporter('test_cache', filename, threshold,
self.nagios_user)
reporter_test.report_and_exit()
except SystemExit, err:
self.assertEqual(
err.code, NAGIOS_EXIT_OK[0],
"Exit with status when the cached data is recent")
def test_threshold(self, message="Hello"):
"""Test the threshold borking mechanism in the reporter."""
message = message.rstrip()
threshold = 1
if message == '':
return
(handle, filename) = tempfile.mkstemp()
os.unlink(filename)
reporter = NagiosReporter('test_cache', filename, threshold, self.nagios_user)
nagios_exit = NAGIOS_EXIT_OK
reporter.cache(nagios_exit, message)
os.close(handle)
try:
reporter_test = NagiosReporter('test_cache', filename, threshold, self.nagios_user)
reporter_test.report_and_exit()
except SystemExit, err:
self.assertEqual(err.code, NAGIOS_EXIT_OK[0],
"Exit with status when the cached data is recent")
nagios_exit = NAGIOS_EXIT_OK
reporter.cache(nagios_exit, message)
os.close(handle)
try:
reporter_test = NagiosReporter('test_cache', filename, threshold, self.nagios_user)
reporter_test.report_and_exit()
except SystemExit, err:
pass
self.assertEqual(err.code, NAGIOS_EXIT_OK[0],
"Exit with status when the cached data is recent")
# restore stdout
buff.close()
sys.stdout = old_stdout
reporter = NagiosReporter('test_cache', filename, threshold, self.nagios_user)
reporter.cache(nagios_exit, message)
time.sleep(threshold + 1)
# redirect stdout
old_stdout = sys.stdout
buff = StringIO.StringIO()
sys.stdout = buff
try:
reporter_test = NagiosReporter('test_cache', filename, threshold, self.nagios_user)
reporter_test.report_and_exit()
except SystemExit, err:
pass
line = buff.getvalue().rstrip()
# restore stdout
buff.close()
def main():
"""Main script"""
options = {
'nagios': ('print out nagios information', None, 'store_true', False, 'n'),
'nagios-check-filename': ('filename of where the nagios check data is stored', str, 'store', NAGIOS_CHECK_FILENAME),
'nagios-check-interval-threshold': ('threshold of nagios checks timing out', None, 'store', NAGIOS_CHECK_INTERVAL_THRESHOLD),
'storage': ('the VSC filesystems that are checked by this script', None, 'extend', []),
'dry-run': ('do not make any updates whatsoever', None, 'store_true', False),
}
opts = simple_option(options)
logger.info('started GPFS quota check run.')
nagios_reporter = NagiosReporter(NAGIOS_HEADER,
opts.options.nagios_check_filename,
opts.options.nagios_check_interval_threshold)
if opts.options.nagios:
nagios_reporter.report_and_exit()
sys.exit(0) # not reached
lockfile = TimestampedPidLockfile(QUOTA_CHECK_LOCK_FILE)
lock_or_bork(lockfile, nagios_reporter)
try:
user_id_map = map_uids_to_names() # is this really necessary?
LdapQuery(VscConfiguration())
gpfs = GpfsOperations()
filesystems = gpfs.list_filesystems().keys()
logger.debug("Found the following GPFS filesystems: %s" % (filesystems))
filesets = gpfs.list_filesets()
logger.debug("Found the following GPFS filesets: %s" % (filesets))
quota = gpfs.list_quota()
for storage in opts.options.storage:
logger.info("Processing quota for storage %s" % (storage))
filesystem = opts.configfile_parser.get(storage, 'filesystem')
if filesystem not in filesystems:
logger.error("Non-existant filesystem %s" % (filesystem))
continue
if filesystem not in quota.keys():
logger.error("No quota defined for storage %s [%s]" % (storage, filesystem))
continue
quota_storage_map = get_mmrepquota_maps(quota[filesystem], storage,filesystem, filesets)
exceeding_filesets = process_fileset_quota(gpfs, storage, filesystem, quota_storage_map['FILESET'])
exceeding_users = process_user_quota(gpfs, storage, filesystem, quota_storage_map['USR'], user_id_map)
logger.warning("storage %s found %d filesets that are exceeding their quota: %s" % (storage,
len(exceeding_filesets),
exceeding_filesets))
logger.warning("storage %s found %d users who are exceeding their quota: %s" % (storage,
len(exceeding_users),
exceeding_users))
notify_exceeding_filesets(gpfs=gpfs,
storage=storage,
filesystem=filesystem,
exceeding_items=exceeding_filesets,
dry_run=opts.options.dry_run)
notify_exceeding_users(gpfs=gpfs,
storage=storage,
filesystem=filesystem,
exceeding_items=exceeding_users,
dry_run=opts.options.dry_run)
sys.exit(1)
except Exception, err:
logger.exception("critical exception caught: %s" % (err))
if not opts.options.dry_run:
nagios_reporter.cache(NAGIOS_EXIT_CRITICAL, NagiosResult("CRITICAL script failed - %s" % (err.message)))
if not opts.options.dry_run:
lockfile.release()
sys.exit(1)
def test_threshold(self, message="Hello"):
"""Test the threshold borking mechanism in the reporter."""
message = message.rstrip()
threshold = 1
if message == '':
return
(handle, filename) = tempfile.mkstemp()
os.unlink(filename)
reporter = NagiosReporter('test_cache', filename, threshold,
self.nagios_user)
# redirect stdout
old_stdout = sys.stdout
buff = StringIO()
sys.stdout = buff
nagios_exit = NAGIOS_EXIT_OK
reporter.cache(nagios_exit, message)
os.close(handle)
raised_exception = None
try:
reporter_test = NagiosReporter('test_cache', filename, threshold,
self.nagios_user)
reporter_test.report_and_exit()
except SystemExit as err:
raised_exception = err
self.assertEqual(raised_exception.code, NAGIOS_EXIT_OK[0],
"Exit with status when the cached data is recent")
# restore stdout
buff.close()
sys.stdout = old_stdout
reporter = NagiosReporter('test_cache', filename, threshold,
self.nagios_user)
reporter.cache(nagios_exit, message)
time.sleep(threshold + 1)
# redirect stdout
old_stdout = sys.stdout
buff = StringIO()
sys.stdout = buff
raised_exception = None
try:
reporter_test = NagiosReporter('test_cache', filename, threshold,
self.nagios_user)
reporter_test.report_and_exit()
except SystemExit as err:
raised_exception = err
line = buff.getvalue().rstrip()
# restore stdout
buff.close()
sys.stdout = old_stdout
self.assertEqual(raised_exception.code, NAGIOS_EXIT_UNKNOWN[0],
"Too old caches lead to unknown status")
self.assertTrue(
line.startswith(
"%s test_cache gzipped JSON file too old (timestamp =" %
(NAGIOS_EXIT_UNKNOWN[1])))
os.unlink(filename)
def main():
# Collect all info
# Note: debug option is provided by generaloption
# Note: other settings, e.g., ofr each cluster will be obtained from the configuration file
options = {
"nagios": ("print out nagios information", None, "store_true", False, "n"),
"nagios_check_filename": (
"filename of where the nagios check data is stored",
str,
"store",
NAGIOS_CHECK_FILENAME,
),
"nagios_check_interval_threshold": (
"threshold of nagios checks timing out",
None,
"store",
NAGIOS_CHECK_INTERVAL_THRESHOLD,
),
"hosts": ("the hosts/clusters that should be contacted for job information", None, "extend", []),
"location": ("the location for storing the pickle file: home, scratch", str, "store", "home"),
"ha": ("high-availability master IP address", None, "store", None),
"dry-run": ("do not make any updates whatsoever", None, "store_true", False),
}
opts = simple_option(options)
if opts.options.debug:
fancylogger.setLogLevelDebug()
nagios_reporter = NagiosReporter(
NAGIOS_HEADER, opts.options.nagios_check_filename, opts.options.nagios_check_interval_threshold
)
if opts.options.nagios:
logger.debug("Producing Nagios report and exiting.")
nagios_reporter.report_and_exit()
sys.exit(0) # not reached
if not proceed_on_ha_service(opts.options.ha):
logger.warning("Not running on the target host in the HA setup. Stopping.")
nagios_reporter.cache(NAGIOS_EXIT_WARNING, NagiosResult("Not running on the HA master."))
sys.exit(NAGIOS_EXIT_WARNING)
lockfile = TimestampedPidLockfile(DCHECKJOB_LOCK_FILE)
lock_or_bork(lockfile, nagios_reporter)
logger.info("Starting dcheckjob")
LdapQuery(VscConfiguration())
clusters = {}
for host in opts.options.hosts:
master = opts.configfile_parser.get(host, "master")
checkjob_path = opts.configfile_parser.get(host, "checkjob_path")
clusters[host] = {"master": master, "path": checkjob_path}
checkjob = Checkjob(clusters, cache_pickle=True, dry_run=True)
(job_information, reported_hosts, failed_hosts) = checkjob.get_moab_command_information()
timeinfo = time.time()
active_users = job_information.keys()
logger.debug("Active users: %s" % (active_users))
logger.debug("Checkjob information: %s" % (job_information))
nagios_user_count = 0
nagios_no_store = 0
for user in active_users:
if not opts.options.dry_run:
try:
(path, store) = get_pickle_path(opts.options.location, user)
user_queue_information = CheckjobInfo({user: job_information[user]})
store(user, path, (timeinfo, user_queue_information))
nagios_user_count += 1
except (UserStorageError, FileStoreError, FileMoveError), _:
logger.error("Could not store pickle file for user %s" % (user))
nagios_no_store += 1
else:
logger.info(
"Dry run, not actually storing data for user %s at path %s"
% (user, get_pickle_path(opts.options.location, user)[0])
)
logger.debug("Dry run, queue information for user %s is %s" % (user, job_information[user]))
def main():
# Collect all info
# Note: debug option is provided by generaloption
# Note: other settings, e.g., ofr each cluster will be obtained from the configuration file
options = {
"nagios": ("print out nagion information", None, "store_true", False, "n"),
"nagios_check_filename": (
"filename of where the nagios check data is stored",
str,
"store",
NAGIOS_CHECK_FILENAME,
),
"nagios_check_interval_threshold": (
"threshold of nagios checks timing out",
None,
"store",
NAGIOS_CHECK_INTERVAL_THRESHOLD,
),
"hosts": ("the hosts/clusters that should be contacted for job information", None, "extend", []),
"information": ("the sort of information to store: user, vo, project", None, "store", "user"),
"location": ("the location for storing the pickle file: gengar, muk", str, "store", "gengar"),
"ha": ("high-availability master IP address", None, "store", None),
"dry-run": ("do not make any updates whatsoever", None, "store_true", False),
}
opts = simple_option(options)
if opts.options.debug:
fancylogger.setLogLevelDebug()
nagios_reporter = NagiosReporter(NAGIOS_HEADER, NAGIOS_CHECK_FILENAME, NAGIOS_CHECK_INTERVAL_THRESHOLD)
if opts.options.nagios:
logger.debug("Producing Nagios report and exiting.")
nagios_reporter.report_and_exit()
sys.exit(0) # not reached
if not proceed_on_ha_service(opts.options.ha):
logger.warning("Not running on the target host in the HA setup. Stopping.")
nagios_reporter.cache(NAGIOS_EXIT_WARNING, NagiosResult("Not running on the HA master."))
sys.exit(NAGIOS_EXIT_WARNING)
lockfile = TimestampedPidLockfile(DSHOWQ_LOCK_FILE)
lock_or_bork(lockfile, nagios_reporter)
logger.info("starting dshowq run")
clusters = {}
for host in opts.options.hosts:
master = opts.configfile_parser.get(host, "master")
showq_path = opts.configfile_parser.get(host, "showq_path")
clusters[host] = {"master": master, "path": showq_path}
showq = Showq(clusters, cache_pickle=True, dry_run=opts.options.dry_run)
(queue_information, reported_hosts, failed_hosts) = showq.get_moab_command_information()
timeinfo = time.time()
active_users = queue_information.keys()
logger.debug("Active users: %s" % (active_users))
logger.debug("Queue information: %s" % (queue_information))
# We need to determine which users should get an updated pickle. This depends on
# - the active user set
# - the information we want to provide on the cluster(set) where this script runs
# At the same time, we need to determine the job information each user gets to see
(target_users, target_queue_information, user_map) = determine_target_information(
opts.options.information, active_users, queue_information
)
nagios_user_count = 0
nagios_no_store = 0
LdapQuery(VscConfiguration())
for user in target_users:
if not opts.options.dry_run:
try:
(path, store) = get_pickle_path(opts.options.location, user)
user_queue_information = target_queue_information[user]
user_queue_information["timeinfo"] = timeinfo
store(user, path, (user_queue_information, user_map[user]))
nagios_user_count += 1
except (UserStorageError, FileStoreError, FileMoveError), err:
logger.error("Could not store pickle file for user %s" % (user))
nagios_no_store += 1
else:
logger.info(
"Dry run, not actually storing data for user %s at path %s"
% (user, get_pickle_path(opts.options.location, user)[0])
)
logger.debug("Dry run, queue information for user %s is %s" % (user, target_queue_information[user]))
nagios_exit = NAGIOS_EXIT_OK
reporter.cache(nagios_exit, message)
os.close(handle)
try:
reporter_test = NagiosReporter('test_cache', filename, threshold, self.nagios_user)
reporter_test.report_and_exit()
except SystemExit, err:
pass
self.assertEqual(err.code, NAGIOS_EXIT_OK[0],
"Exit with status when the cached data is recent")
# restore stdout
buff.close()
sys.stdout = old_stdout
reporter = NagiosReporter('test_cache', filename, threshold, self.nagios_user)
reporter.cache(nagios_exit, message)
time.sleep(threshold + 1)
# redirect stdout
old_stdout = sys.stdout
buff = StringIO.StringIO()
sys.stdout = buff
try:
reporter_test = NagiosReporter('test_cache', filename, threshold, self.nagios_user)
reporter_test.report_and_exit()
except SystemExit, err:
pass
line = buff.getvalue().rstrip()
# restore stdout
buff.close()
def main():
"""The main."""
# Note: debug option is provided by generaloption
# Note: other settings, e.g., ofr each cluster will be obtained from the configuration file
options = {
'nagios': ('print out nagios information', None, 'store_true', False, 'n'),
'nagios-check-filename': ('filename of where the nagios check data is stored',
str,
'store',
NAGIOS_CHECK_FILENAME),
'nagios-check-interval-threshold': ('threshold of nagios checks timing out',
None,
'store',
NAGIOS_CHECK_INTERVAL_THRESHOLD),
'location': ('path to store the gzipped files', None, 'store', QUOTA_LOG_ZIP_PATH),
'ha': ('high-availability master IP address', None, 'store', None),
'dry-run': ('do not make any updates whatsoever', None, 'store_true', False),
}
opts = simple_option(options)
nagios_reporter = NagiosReporter(NAGIOS_HEADER,
opts.options.nagios_check_filename,
opts.options.nagios_check_interval_threshold)
if opts.options.nagios:
logger.debug("Producing Nagios report and exiting.")
nagios_reporter.report_and_exit()
sys.exit(0) # not reached
if not proceed_on_ha_service(opts.options.ha):
logger.warning("Not running on the target host in the HA setup. Stopping.")
nagios_reporter.cache(NAGIOS_EXIT_WARNING,
NagiosResult("Not running on the HA master."))
sys.exit(NAGIOS_EXIT_WARNING)
lockfile = TimestampedPidLockfile(QUOTA_LOG_LOCK_FILE)
lock_or_bork(lockfile, nagios_reporter)
logger.info("starting quota_log run")
filesystem_error = 0
filesystem_ok = 0
error = False
try:
gpfs = GpfsOperations()
quota = gpfs.list_quota()
for key in quota:
try:
filename = "gpfs_quota_%s_%s.gz" % (time.strftime("%Y%m%d-%H:%M"), key)
path = os.path.join(opts.options.location, filename)
zipfile = gzip.open(path, 'wb', 9) # Compress to the max
zipfile.write(json.dumps(quota[key]))
zipfile.close()
filesystem_ok += 1
logger.info("Stored quota information for FS %s" % (key))
except Exception, err:
logger.exception("Failed storing quota information for FS %s" % (key))
filesystem_error += 1
except Exception, err:
logger.exception("Failure obtaining GPFS quota")
error = True
