def run(self):
''' Main function that runs the check '''
self.parse_args()
self.metric_sender = MetricSender(verbose=self.args.verbose,
debug=self.args.debug)
self.oc = OCUtil(namespace='openshift-infra',
config_file='/tmp/admin.kubeconfig',
verbose=self.args.verbose)
pod_report = self.check_pods()
self.get_hawkular_creds()
metrics_report = self.check_node_metrics()
# if metrics_report = 0, we need this check run again
if metrics_report['success'] == 0:
# sleep for 5 seconds, then run the second time node check
logger.info(
"The first time metrics check failed, 5 seconds later will start a second time check"
)
time.sleep(commandDelay)
logger.info("starting the second time metrics check")
metrics_report = self.check_node_metrics()
# persist second attempt if fails
if metrics_report['success'] == 0:
self.persist_details(metrics_report)
self.report_to_zabbix(pod_report, metrics_report['success'])
def __init__(
self,
args=None,
):
'''initial for the InfraNodePodStatus'''
self.args = args
self.kubeconfig = '/tmp/admin.kubeconfig'
self.oc = OCUtil(namespace=self.args.namespace,
config_file=self.kubeconfig)
def run(self):
''' Main function that runs the check '''
self.parse_args()
self.zagg_sender = ZaggSender(verbose=self.args.verbose, debug=self.args.debug)
self.oc = OCUtil(namespace='openshift-infra', config_file=self.kubeconfig, verbose=self.args.verbose)
pod_report = self.check_pods()
metrics_report = self.check_node_metrics()
self.report_to_zabbix(pod_report, metrics_report)
def run(self):
""" Main function that runs the check """
self.parse_args()
self.metric_sender = MetricSender(verbose=self.args.verbose,
debug=self.args.debug)
self.oc = OCUtil(namespace=self.get_logging_namespace(),
config_file='/tmp/admin.kubeconfig',
verbose=self.args.verbose)
self.get_pods()
oldest_buffer = self.check_fluentd_queues()
self.send_metrics(oldest_buffer)
def run(self):
"""Main function to run the check"""
self.ocutil = OCUtil(config_file=self.kubeconfig,
verbose=self.args.verbose)
self.zgs = ZaggSender(verbose=self.args.verbose, debug=self.args.debug)
self.check_all_router_health()
self.check_router_replica_count()
if self.args.dry_run:
self.zgs.print_unique_metrics_key_value()
else:
self.zgs.send_metrics()
def __init__(self):
""" initialize EBSStuckVolumesCheck class """
self.args = None
self.vol_state_data = None
self.parse_args()
# Make sure we're using the profile they've requested.
if self.args.aws_creds_profile:
os.environ['AWS_PROFILE'] = self.args.aws_creds_profile
self.eu = EbsUtil(self.args.region, verbose=self.args.verbose)
self.ocutil = OCUtil(verbose=self.args.verbose)
self.mts = MetricSender(verbose=self.args.verbose)
def main():
''' Gather and send details on all visible S3 buckets '''
logger.info("start")
discovery_key = "disc.aws"
discovery_macro = "#S3_BUCKET"
prototype_s3_size = "disc.aws.size"
prototype_s3_count = "disc.aws.objects"
args = parse_args()
if args.verbose:
logger.setLevel(logging.DEBUG)
logger.debug("verbose flag set")
ocutil = OCUtil()
dc_yaml = ocutil.get_dc('docker-registry')
registry_config_secret = get_registry_config_secret(dc_yaml)
oc_yaml = ocutil.get_secrets(registry_config_secret)
aws_access, aws_secret = get_aws_creds(oc_yaml)
awsutil = AWSUtil(aws_access, aws_secret, args.debug)
bucket_list = awsutil.get_bucket_list(args.debug)
bucket_stats = {}
for bucket in bucket_list:
s3_size, s3_objects = awsutil.get_bucket_info(bucket, args.debug)
bucket_stats[bucket] = {"size": s3_size, "objects": s3_objects}
if args.debug:
print "Bucket stats: " + str(bucket_stats)
if args.test:
print "Test-only. Received results: " + str(bucket_stats)
else:
zgs = ZaggSender(verbose=args.debug)
zgs.add_zabbix_dynamic_item(discovery_key, discovery_macro,
bucket_list)
for bucket in bucket_stats.keys():
zab_key = "{}[{}]".format(prototype_s3_size, bucket)
zgs.add_zabbix_keys(
{zab_key: int(round(bucket_stats[bucket]["size"]))})
zab_key = "{}[{}]".format(prototype_s3_count, bucket)
zgs.add_zabbix_keys({zab_key: bucket_stats[bucket]["objects"]})
zgs.send_metrics()
def run(self):
''' Main function that runs the check '''
self.parse_args()
self.metric_sender = MetricSender(verbose=self.args.verbose,
debug=self.args.debug)
self.oc = OCUtil(namespace='openshift-infra',
config_file='/tmp/admin.kubeconfig',
verbose=self.args.verbose)
pod_report = self.check_pods()
self.get_hawkular_creds()
metrics_report = self.check_node_metrics()
self.report_to_zabbix(pod_report, metrics_report)
def get_registry_config_secret(yaml_results):
''' Find the docker registry config secret '''
ocutil = OCUtil()
volumes = yaml_results['spec']['template']['spec']['volumes']
for volume in volumes:
if 'emptyDir' in volume:
continue
secret_dict = ocutil.get_secrets(volume['secret']['secretName'])
if 'config.yml' in secret_dict['data']:
return volume['secret']['secretName']
print "Unable to find the %s the docker registry config"
print "Please run \"oc get dc docker-registry\" to investigate"
sys.exit(1)
def get_logging_namespace(self):
""" Determine which logging namespace is in use """
# Assume the correct namespace is 'openshift-logging' and fall back to 'logging'
# if that assumption ends up being wrong.
oc_client = OCUtil(namespace='openshift-logging',
config_file='/tmp/admin.kubeconfig',
verbose=self.args.verbose)
logger.info("Determining which namespace is in use...")
try:
oc_client.get_dc('logging-kibana')
# If the previous call didn't throw an exception, logging is deployed in this namespace.
logger.info("Using namespace: openshift-logging")
return 'openshift-logging'
except subprocess.CalledProcessError:
logger.info("Using namespace: logging")
return 'logging'
def main():
''' main() '''
args = parse_args()
if args.verbose:
logger.setLevel(logging.DEBUG)
logger.info("Starting")
# TODO: include this in library
projects_info = OCUtil().get_projects()
maxDelta = testProjects(
projects_info['items'],
current_time=datetime.datetime.now(),
)
send_zagg_data(maxDelta)
logger.info('Oldest Terminating project: %s seconds', maxDelta)
def main():
''' main() '''
args = parse_args()
if args.verbose:
logger.setLevel(logging.DEBUG)
logger.info("Starting")
# TODO: include this in library
projects_info = OCUtil()._run_cmd("oc get projects -o yaml")
time_keeps_max = testProjects(
projects_info['items'],
current_time=datetime.datetime.now(),
)
send_zagg_data(time_keeps_max)
logger.info('Oldest Terminating project: %s seconds', time_keeps_max)
def main():
''' Gather and send details on all visible S3 buckets '''
discovery_key = "disc.gcp"
discovery_macro = "#GCS_BUCKET"
prototype_bucket_size = "disc.gcp.size"
prototype_bucket_count = "disc.gcp.objects"
args = parse_args()
ocutil = OCUtil()
dc_yaml = ocutil.get_dc('docker-registry')
registry_config_secret = get_registry_config_secret(dc_yaml)
oc_yaml = ocutil.get_secrets(registry_config_secret)
bucket = get_gcp_info(oc_yaml)
gsutil = GcloudUtil(verbose=args.debug)
bucket_list = gsutil.get_bucket_list()
bucket_stats = {}
for bucket in bucket_list:
size, objects = gsutil.get_bucket_info(bucket)
bucket_stats[bucket] = {"size": size, "objects": objects}
if args.debug:
print "Bucket stats: " + str(bucket_stats)
if args.test:
print "Test-only. Received results: " + str(bucket_stats)
else:
zgs = ZaggSender(verbose=args.debug)
zgs.add_zabbix_dynamic_item(discovery_key, discovery_macro, bucket_list)
for bucket in bucket_stats.keys():
zab_key = "{}[{}]".format(prototype_bucket_size, bucket)
zgs.add_zabbix_keys({zab_key: int(round(bucket_stats[bucket]["size"]))})
zab_key = "{}[{}]".format(prototype_bucket_count, bucket)
zgs.add_zabbix_keys({zab_key: bucket_stats[bucket]["objects"]})
zgs.send_metrics()
def run(self):
""" Main function to run the check """
self.parse_args()
self.get_kubeconfig()
ocutil = OCUtil(config_file=self.kubeconfig, verbose=self.args.verbose)
self.metric_sender = MetricSender(verbose=self.args.verbose, debug=self.args.debug)
try:
oc_yaml = ocutil.get_service('docker-registry')
self.get_registry_service(oc_yaml)
oc_yaml = ocutil.get_endpoint('docker-registry')
self.get_registry_endpoints(oc_yaml)
except Exception as ex:
print "Problem retreiving registry IPs: %s " % ex.message
self.registry_service_check()
self.registry_health_check()
self.metric_sender.send_metrics()
def main():
''' Gather and send details on all visible S3 buckets '''
#get the region
with open('/container_setup/monitoring-config.yml', 'r') as f:
doc = yaml.load(f)
bucket_region = doc['oso_region']
args = parse_args()
ocutil = OCUtil()
dc_yaml = ocutil.get_dc('docker-registry')
registry_config_secret = get_registry_config_secret(dc_yaml)
oc_yaml = ocutil.get_secrets(registry_config_secret)
aws_access, aws_secret = get_aws_creds(oc_yaml)
awsutil = AWSUtil(aws_access, aws_secret, args.debug)
bucket_list = awsutil.get_bucket_list(verbose=args.debug,
BucketRegion=bucket_region)
bucket_stats = {}
for bucket in bucket_list:
#print bucket
s3_size, s3_objects = awsutil.get_bucket_info(
bucket, verbose=args.debug, BucketRegion=bucket_region)
bucket_stats[bucket] = {"size": s3_size, "objects": s3_objects}
if args.debug:
print "Bucket stats: " + str(bucket_stats)
if args.test:
print "Test-only. Received results: " + str(bucket_stats)
else:
send_zagg_data(bucket_list, bucket_stats, args)
def __init__(self):
'''initial for the InfraNodePodStatus'''
self.kubeconfig = '/tmp/admin.kubeconfig'
self.oc = OCUtil(namespace='default', config_file=self.kubeconfig)
# Our jenkins server does not include these rpms.
# In the future we might move this to a container where these
# libs might exist
#pylint: disable=import-error
from openshift_tools.monitoring.ocutil import OCUtil
from openshift_tools.monitoring.metric_sender import MetricSender
import logging
logging.basicConfig(
format='%(asctime)s - %(relativeCreated)6d - %(levelname)-8s - %(message)s',
)
logger = logging.getLogger()
logger.setLevel(logging.INFO)
ocutil = OCUtil()
valid_build_states = ["cancelled", "complete", "new", "error", "failed"]
def runOCcmd(cmd, base_cmd='oc'):
""" log commands through ocutil """
logger.info(base_cmd + " " + cmd)
return ocutil.run_user_cmd(
cmd,
base_cmd=base_cmd,
)
def parse_args():
""" parse the args from the cli """
