cyberes
/
icinga2-checks
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

check_graylog: fix

This commit is contained in:
Cyberes 2023-06-27 18:28:49 -06:00
parent 725b7a2bab
commit 61ec9a8580
1 changed files with 33 additions and 25 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

58
check_graylog.py
View File

@ -41,7 +41,8 @@ def main():
parser.add_argument('--insecure', action='store_false', help="Don't verify SSL")
parser.add_argument('--crit-notif', action='store_true', help='Return critical when there are notifications')
parser.add_argument('--html', action='store_true', help='Print HTML')
parser.add_argument('--cluster-metrics', action='store_true', help='Also gather cluster metrics and check for notifications')
parser.add_argument('--cluster-metrics', action='store_true',
help='Also gather cluster metrics and check for notifications')
args = parser.parse_args()
base_url = args.url.strip('/')
@ -72,8 +73,25 @@ def main():
}
if args.input:
# Get the basic input metadata
input_data = transform_inputs(fetch_with_retry(f'{base_url}/api/cluster/inputstates', headers=headers, auth=(args.token, 'token'), verify=args.insecure).json()).get(args.input, {})
input_data = transform_inputs(
fetch_with_retry(f'{base_url}/api/cluster/inputstates', headers=headers, auth=(args.token, 'token'),
verify=args.insecure).json()).get(args.input, {})
# Get it over with
if bool(input_data) and input_data.get('state') == 'RUNNING':
input_name = input_data["message_input"]["title"]
text_result = f'Graylog input "{input_name}" is running.'
else:
input_name = args.input
if args.html:
text_result = f'Graylog input <a href="{base_url}/system/inputs" target="_blank">"{input_name}" is not running!</a>'
else:
text_result = f'Graylog input "{input_name}" is not running!'
print_icinga2_check_status(text_result, nagios.STATE_CRIT)
sys.exit(nagios.STATE_CRIT)
# If the input is running, continue gathering metrics and other health checks
input_name_clean = input_name.lower().replace(' ', '_').replace('-', '_')
type = input_data['message_input']['type']
metrics_json['metrics'] = metrics_json['metrics'] + [
f'{type}.{args.input}.incomingMessages',
@ -85,7 +103,8 @@ def main():
f'{type}.{args.input}.read_bytes_total',
]
r = fetch_with_retry(f'{base_url}/api/cluster/metrics/multiple', method='post', headers=headers, auth=(args.token, 'token'),
r = fetch_with_retry(f'{base_url}/api/cluster/metrics/multiple', method='post', headers=headers,
auth=(args.token, 'token'),
verify=args.insecure, json=metrics_json).json()
input_metrics = r[list(r.keys())[0]]['metrics']
@ -111,22 +130,6 @@ def main():
perfdata = {}
if args.input:
# Get it over with
if bool(input_data) and input_data.get('state') == 'RUNNING':
input_name = input_data["message_input"]["title"]
text_result = f'Graylog input "{input_name}" is running.'
else:
input_name = args.input
if args.html:
text_result = f'Graylog input <a href="{base_url}/system/inputs" target="_blank">"{input_name}" is not running!</a>'
else:
text_result = f'Graylog input "{input_name}" is not running!'
print_icinga2_check_status(text_result, nagios.STATE_CRIT)
sys.exit(nagios.STATE_CRIT)
# If the input is running, continue gathering metrics and other health checks
input_name_clean = input_name.lower().replace(' ', '_').replace('-', '_')
# Some metric names are changed for better readability
perfdata.update({
f'{input_name_clean}_incoming_messages_rate_per_sec_1min': {
@ -164,15 +167,18 @@ def main():
jvm_mem_usage_crit = int(metrics_data['jvm_memory_heap_max'] / int(100 / args.crit_mem))
# Get traffic data for last 24 hrs
traffic_last_24_hrs = fetch_with_retry(f'{base_url}/api/system/cluster/traffic?daily=false', headers=headers, auth=(args.token, 'token'), verify=args.insecure).json()
traffic_last_24_hrs = fetch_with_retry(f'{base_url}/api/system/cluster/traffic?daily=false', headers=headers,
auth=(args.token, 'token'), verify=args.insecure).json()
input_traffic_avg = sum([v for k, v in traffic_last_24_hrs['input'].items() if parse_traffic_ts(k)])
output_traffic_avg = sum([v for k, v in traffic_last_24_hrs['output'].items() if parse_traffic_ts(k)])
elasticsearch_health = fetch_with_retry(f'{base_url}/api/system/indexer/cluster/health', headers=headers, auth=(args.token, 'token'), verify=args.insecure).json()
elasticsearch_health = fetch_with_retry(f'{base_url}/api/system/indexer/cluster/health', headers=headers,
auth=(args.token, 'token'), verify=args.insecure).json()
elasticsearch_status = elasticsearch_health['status'].lower()
elasticsearch_active_shards = elasticsearch_health['shards']['active']
indexer_failures = fetch_with_retry(f'{base_url}/api/system/indexer/failures?limit=10&offset=0', headers=headers, auth=(args.token, 'token'), verify=args.insecure).json()
indexer_failures = fetch_with_retry(f'{base_url}/api/system/indexer/failures?limit=10&offset=0',
headers=headers, auth=(args.token, 'token'), verify=args.insecure).json()
perfdata.update({
'throughput_input_1_sec_rate': {
@ -219,7 +225,8 @@ def main():
# Check for notifications
if args.cluster_metrics:
notifications = fetch_with_retry(f'{base_url}/api/system/notifications', headers=headers, auth=(args.token, 'token'), verify=args.insecure).json()
notifications = fetch_with_retry(f'{base_url}/api/system/notifications', headers=headers,
auth=(args.token, 'token'), verify=args.insecure).json()
if notifications['total'] > 0:
notif_str = 'Notifications:'
for notification in notifications['notifications']:
@ -249,7 +256,8 @@ def main():
print_icinga2_check_status(f'unknown Elasticsearch health: {elasticsearch_status}', nagios.STATE_UNKNOWN)
sys.exit(nagios.STATE_UNKNOWN)
jvm_mem_usage_state = get_state(int(metrics_data['jvm_memory_heap_used']), jvm_mem_usage_warn, jvm_mem_usage_crit, operator='gt')
jvm_mem_usage_state = get_state(int(metrics_data['jvm_memory_heap_used']), jvm_mem_usage_warn,
jvm_mem_usage_crit, operator='gt')
if jvm_mem_usage_state != nagios.STATE_OK:
text_result += f' JVM memory usage is high!'