def get_queue_state(self):
q = StateSetMetricFamily('fah_queue_state',
'State of the queue',
labels=['id', 'slot', 'core', 'unit'])
states = {
'UNKNOWN': False,
'PAUSED': False,
'UPDATING': False,
'FINISHED': False,
'UPLOADED': False,
'RUNNING': False,
'ACTIVE': False,
'ONLINE': False,
'FINISHING': False,
'FAILED': False,
'ERROR': False,
'FAULTY': False,
'SHUTDOWN': False,
'CONNECTING': False,
'OFFLINE': False,
'READY': False,
'DUMP': False,
'DOWNLOAD': False
}
for queue in self.queue_info:
queue_states = states
queue_states[queue['state']] = True
q.add_metric(
[queue['id'], queue['slot'], queue['core'], queue['unit']],
queue_states)
yield q
def get_metric_authentication(self):
""" Shows if the current run was authenticated """
m = StateSetMetricFamily(
'authentication',
'Shows if authentication is enabled',
labels=['exchange'],
)
try:
m.add_metric(
[f'{self.exchange.exchange}'],
{'enabled': self.exchange.get_enable_authentication()},
)
except AttributeError:
pass
return m
def get_status_metrics(self, result):
metric = StateSetMetricFamily(
'dkron_job_status',
'Dkron job status',
labels=["jobname"])
for job in result:
name = job['name']
# If there's a null result, we want to export a zero.
status = True
status_txt = job.get('status') or 'failed'
if status_txt == 'failed':
status = False
metric.add_metric([name], {'success': status})
# job_status = job['status']
return metric
def test_stateset_labels(self):
cmf = StateSetMetricFamily('s', 'help', labels=['foo'])
cmf.add_metric(['bar'], {
'a': False,
'b': False,
})
self.custom_collector(cmf)
self.assertEqual(
0, self.registry.get_sample_value('s', {
'foo': 'bar',
's': 'a'
}))
self.assertEqual(
0, self.registry.get_sample_value('s', {
'foo': 'bar',
's': 'b'
}))
def get_schedule_status_metrics(self, result):
metric = StateSetMetricFamily(
'dkron_job_schedule_status',
'Dkron job schedule status',
labels=["jobname"])
for job in result:
name = job['name']
# If there's a null result, we want to export a zero.
next_date_str = job.get('next') or '2020-01-01T00:00:00.000Z'
next_date = dateutil.parser.parse(next_date_str)
diff_date = next_date - datetime.datetime.now(datetime.timezone.utc)
if diff_date < datetime.timedelta(minutes=-1):
metric.add_metric([name], {'success': False})
else:
metric.add_metric([name], {'success': True})
# job_status = job['status']
return metric
def collect(self):
license_counter = CounterMetricFamily("active_licenses",
"active license count",
labels=("license", ))
client_state = StateSetMetricFamily("active_licenses",
"active license count",
labels=("host", "port", "license",
"user"))
for license in self._licenses:
license_counter.add_metric([license.name], len(license.licenses))
for client in license.licenses:
client_state.add_metric(
(client.host, str(client.port), license.name, client.user),
value={"active": True})
yield (license_counter)
yield (client_state)
def collect(self):
# get SID list from os dir
sidList = get_sid_list()
for sid in sidList:
c = consul.Consul(host=os.environ.get('CONSUL_HOST') if
os.environ.get('CONSUL_HOST') else '127.0.0.1',
port=23345,
scheme='http')
kvid, kvv = c.kv.get(sid + '_login')
if kvv:
# get SID login info from consul
kvvDict = json.loads(kvv['Value'])
conn = None
for instance in get_instance_list_by_sid(sid):
if not conn and instance['type'] == 'DIALOG':
conn = R3rfcconn(r3ashost='127.0.0.1',
r3sysnr=instance['sysnr'],
r3client=kvvDict['r3client'],
r3user=kvvDict['r3user'],
r3pwd=kvvDict['r3pwd'])
instance_check_cmd = f'su - {sid.lower()}adm -c "sapcontrol -nr {instance["sysnr"]} -function GetProcessList"'
instance_check_cmd_args = shlex.split(instance_check_cmd)
sp = subprocess.run(instance_check_cmd_args,
capture_output=True)
output = sp.stdout.decode('utf-8')
outputlines = output.splitlines()
if 'Red' in output:
g_instancestatus = StateSetMetricFamily(
"InstanceStatus",
'Instance Status Check in SID',
labels=['SID', 'Instance'])
g_instancestatus.add_metric([sid, instance["profile"]],
{'status': False})
yield g_instancestatus
else:
g_instancestatus = StateSetMetricFamily(
"InstanceStatus",
'Instance Status Check in SID',
labels=['SID', 'Instance'])
g_instancestatus.add_metric([sid, instance["profile"]],
{'status': True})
yield g_instancestatus
pass
pass
if conn:
for p in get_instance_servername_list_by_sid(sid):
# master identification
servername = p['servername']
profile = p['profile']
kvid_master, kvv_master = c.kv.get(sid + '_master')
if kvv_master:
kvvDict_master = json.loads(kvv_master['Value'])
if servername == kvvDict_master['servername']:
# during user count, by user type
USRLIST = conn.get_user_list()
g_usercount = GaugeMetricFamily(
"UserCount",
'System Overall User Count',
labels=['SID'])
g_usercount.add_metric([sid], len(USRLIST))
yield g_usercount
# during dump count
DUMPLIST = conn.get_dump_list()
g_dumpcount = GaugeMetricFamily(
"DumpCount",
'System Overall Dump Count',
labels=['SID'])
g_dumpcount.add_metric([sid], len(DUMPLIST))
yield g_dumpcount
# get bk job status
job_status = conn.get_bkjob_status_count()
g_jobstatus = GaugeMetricFamily(
"BKJobCount",
'Current Background Job Count Status',
labels=['SID', 'BKJobStatus'])
g_jobstatus.add_metric([sid, 'Finish'],
job_status['finish'])
g_jobstatus.add_metric([sid, 'Running'],
job_status['running'])
g_jobstatus.add_metric([sid, 'Cancel'],
job_status['cancel'])
yield g_jobstatus
else:
c.kv.put(sid + '_master',
json.dumps({"servername": servername}))
# during user count, by user type
USRLIST = conn.get_user_list()
g_usercount = GaugeMetricFamily(
"UserCount",
'System Overall User Count',
labels=['SID'])
g_usercount.add_metric([sid], len(USRLIST))
yield g_usercount
# during dump count
DUMPLIST = conn.get_dump_list()
g_dumpcount = GaugeMetricFamily(
"DumpCount",
'System Overall Dump Count',
labels=['SID'])
g_dumpcount.add_metric([sid], len(DUMPLIST))
yield g_dumpcount
# get bk job status
job_status = conn.get_bkjob_status_count()
g_jobstatus = GaugeMetricFamily(
"BKJobCount",
'Current Background Job Count Status',
labels=['SID', 'BKJobStatus'])
g_jobstatus.add_metric([sid, 'Finish'],
job_status['finish'])
g_jobstatus.add_metric([sid, 'Running'],
job_status['running'])
g_jobstatus.add_metric([sid, 'Cancel'],
job_status['cancel'])
yield g_jobstatus
# during workprocess count, by wp type
wplist = conn.get_server_wp_list(servername)
running_dia_count = 0
running_upd_count = 0
running_btc_count = 0
for wp in wplist:
if wp['WP_ISTATUS'] != 2:
if wp['WP_TYP'] == 'DIA':
running_dia_count += 1
pass
if wp['WP_TYP'] == 'BTC':
running_btc_count += 1
pass
if wp['WP_TYP'] == 'UPD':
running_upd_count += 1
pass
g_wpcount = GaugeMetricFamily(
"WorkprocessCount",
'WorkprocessCount of One Instance in SID group by Type',
labels=['SID', 'Instance', 'WorkprocessType'])
g_wpcount.add_metric([sid, profile, 'DIA'],
running_dia_count)
g_wpcount.add_metric([sid, profile, 'BTC'],
running_btc_count)
g_wpcount.add_metric([sid, profile, 'UPD'],
running_upd_count)
yield g_wpcount
# st02 data by instance
st02data = conn.get_st02_data()
st02data['instance'] = profile
g_st02_sapmemory = GaugeMetricFamily(
"SAPMemory",
'SAP Memory Current Use % in TCode ST02',
labels=[
'SID', 'Instance', 'SAPMemoryCurrentUsePercent'
])
g_st02_sapmemory.add_metric(
[sid, profile, 'PageArea'],
round(
float(st02data['PAGING_AREA']['CURR_USED']) /
float(st02data['PAGING_AREA']['AREA_SIZE']), 4)
* 100)
g_st02_sapmemory.add_metric(
[sid, profile, 'ExtendedMemory'],
round(
float(
st02data['EXTENDED_MEMORY_USAGE']['USED'])
/ float(st02data['EXTENDED_MEMORY_USAGE']
['TOTAL']), 4) * 100)
g_st02_sapmemory.add_metric(
[sid, profile, 'HeapMemory'],
round(
float(st02data['HEAP_MEMORY_USAGE']['USED']) /
float(st02data['HEAP_MEMORY_USAGE']['TOTAL']),
4) * 100)
yield g_st02_sapmemory
g_st02_callstatics = GaugeMetricFamily(
"CallStatistics",
'SAP Call Statistics HitRadio % in TCode ST02',
labels=[
'SID', 'Instance',
'CallStatisticsHitRadioPercent'
])
g_st02_callstatics.add_metric(
[sid, profile, 'DIRECT'],
st02data['TOTAL_HITRATIO']['DIRECT'])
g_st02_callstatics.add_metric(
[sid, profile, 'SEQUENTIAL'],
st02data['TOTAL_HITRATIO']['SEQUENTIAL'])
g_st02_callstatics.add_metric(
[sid, profile, 'AVERAGE'],
st02data['TOTAL_HITRATIO']['AVERAGE'])
yield g_st02_callstatics
# es = Elasticsearch(hosts=[{
# "host": "{master_ip}",
# "port": 23392
# }])
# es.index(
# index="ST02",
# body=st02data,
# )
# st03 summary data by instance
st03detail_summary = conn.get_st03_data_summary()
st03detail = {}
st03detail['instance'] = profile
st03detail['SUMMARY'] = st03detail_summary
g_st03detail_steps = GaugeMetricFamily(
"DialogSteps",
'Dialog Steps in TCode ST03',
labels=['SID', 'Instance', 'DialogSteps'])
g_st03detail_averageRespTime = GaugeMetricFamily(
"AverageRespTime",
'Average Resp Time in TCode ST03',
labels=['SID', 'Instance', 'AverageRespTime'])
g_st03detail_averageDBTime = GaugeMetricFamily(
"AverageDBTime",
'Average DB Time in TCode ST03',
labels=['SID', 'Instance', 'AverageDBTime'])
for sumary in st03detail_summary:
g_st03detail_steps.add_metric(
[sid, profile, sumary['TASKTYPE']],
float(sumary['COUNT']))
g_st03detail_averageRespTime.add_metric(
[sid, profile, sumary['TASKTYPE']],
round(
float(sumary['RESPTI']) /
float(sumary['COUNT'])))
# g_st03detail_averageDBTime.add_metric(
# [sid, profile, sumary['TASKTYPE']],
# round(sumary['RESPTI'] / sumary['COUNT']))
yield g_st03detail_steps
yield g_st03detail_averageRespTime
# es.index(
# index="ST03",
# body=st03detail,
# )
conn.close()
'''
def collect(self):
# get SID list from os dir
sidList = get_sid_list()
for sid in sidList:
c = consul.Consul(host=os.environ.get('CONSUL_HOST') if
os.environ.get('CONSUL_HOST') else '127.0.0.1',
port=23345,
scheme='http')
kvid, kvv = c.kv.get(sid + '_login')
if kvv:
# get SID login info from consul
kvvDict = json.loads(kvv['Value'])
conn = None
for instance in get_instance_list_by_sid(sid):
if not conn and instance['type'] == 'DIALOG':
conn = R3rfcconn(r3ashost='127.0.0.1',
r3sysnr=instance['sysnr'],
r3client=kvvDict['r3client'],
r3user=kvvDict['r3user'],
r3pwd=kvvDict['r3pwd'])
instance_check_cmd = f'su - {sid.lower()}adm -c "sapcontrol -nr {instance["sysnr"]} -function GetProcessList"'
instance_check_cmd_args = shlex.split(instance_check_cmd)
sp = subprocess.run(instance_check_cmd_args,
capture_output=True)
output = sp.stdout.decode('utf-8')
outputlines = output.splitlines()
if 'Red' in output:
g_instancestatus = StateSetMetricFamily(
"InstanceStatus",
'Instance Status Check in SID',
labels=['SID', 'Instance'])
g_instancestatus.add_metric([sid, instance["profile"]],
{'status': False})
yield g_instancestatus
else:
g_instancestatus = StateSetMetricFamily(
"InstanceStatus",
'Instance Status Check in SID',
labels=['SID', 'Instance'])
g_instancestatus.add_metric([sid, instance["profile"]],
{'status': True})
yield g_instancestatus
pass
pass
if conn:
for p in get_instance_servername_list_by_sid(sid):
# master identification
servername = p['servername']
profile = p['profile']
kvid_master, kvv_master = c.kv.get(sid + '_master')
if kvv_master:
kvvDict_master = json.loads(kvv_master['Value'])
if servername == kvvDict_master['servername']:
# during user count, by user type
USRLIST = conn.get_user_list()
g_usercount = GaugeMetricFamily(
"UserCount",
'System Overall User Count',
labels=['SID'])
g_usercount.add_metric([sid], len(USRLIST))
yield g_usercount
# during dump count
DUMPLIST = conn.get_dump_list()
g_dumpcount = GaugeMetricFamily(
"DumpCount",
'System Overall Dump Count',
labels=['SID'])
g_dumpcount.add_metric([sid], len(DUMPLIST))
yield g_dumpcount
# get bk job status
job_status = conn.get_bkjob_status_count()
g_jobstatus = GaugeMetricFamily(
"BKJobCount",
'Current Background Job Count Status',
labels=['SID', 'BKJobStatus'])
g_jobstatus.add_metric([sid, 'Finish'],
job_status['finish'])
g_jobstatus.add_metric([sid, 'Running'],
job_status['running'])
g_jobstatus.add_metric([sid, 'Cancel'],
job_status['cancel'])
yield g_jobstatus
else:
c.kv.put(sid + '_master',
json.dumps({"servername": servername}))
# during user count, by user type
USRLIST = conn.get_user_list()
g_usercount = GaugeMetricFamily(
"UserCount",
'System Overall User Count',
labels=['SID'])
g_usercount.add_metric([sid], len(USRLIST))
yield g_usercount
# during dump count
DUMPLIST = conn.get_dump_list()
g_dumpcount = GaugeMetricFamily(
"DumpCount",
'System Overall Dump Count',
labels=['SID'])
g_dumpcount.add_metric([sid], len(DUMPLIST))
yield g_dumpcount
# get bk job status
job_status = conn.get_bkjob_status_count()
g_jobstatus = GaugeMetricFamily(
"BKJobCount",
'Current Background Job Count Status',
labels=['SID', 'BKJobStatus'])
g_jobstatus.add_metric([sid, 'Finish'],
job_status['finish'])
g_jobstatus.add_metric([sid, 'Running'],
job_status['running'])
g_jobstatus.add_metric([sid, 'Cancel'],
job_status['cancel'])
yield g_jobstatus
# during workprocess count, by wp type
wplist = conn.get_server_wp_list(servername)
running_dia_count = 0
running_upd_count = 0
running_btc_count = 0
for wp in wplist:
if wp['WP_ISTATUS'] != 2:
if wp['WP_TYP'] == 'DIA':
running_dia_count += 1
pass
if wp['WP_TYP'] == 'BTC':
running_btc_count += 1
pass
if wp['WP_TYP'] == 'UPD':
running_upd_count += 1
pass
g_wpcount = GaugeMetricFamily(
"WorkprocessCount",
'WorkprocessCount of One Instance in SID group by Type',
labels=['SID', 'Instance', 'WorkprocessType'])
g_wpcount.add_metric([sid, profile, 'DIA'],
running_dia_count)
g_wpcount.add_metric([sid, profile, 'BTC'],
running_btc_count)
g_wpcount.add_metric([sid, profile, 'UPD'],
running_upd_count)
yield g_wpcount
conn.close()
'''