def check_health(self):
imp_map = {'OK': 'GREEN', 'WARNING': 'YELLOW', 'CRITICAL': 'RED'}
other_map = {'OK': 'GREEN'}
color_counter = collections.Counter()
important_services = self.get_important_services()
for alert in self.provider.get_alerts_data(self.service):
alert_summary = alert.get('state', 'UNKNOWN')
if self.service in important_services:
target = imp_map.get(alert_summary, 'RED')
else:
target = other_map.get(alert_summary, 'YELLOW')
color_counter[target] += 1
if color_counter['RED'] > 0 and color_counter['YELLOW'] > 0:
raise health_check_base.RedHealthError(
_("Ambari Monitor has responded that cluster has "
"%(red)d critical and %(yellow)d warning alert(s)") % {
'red': color_counter['RED'],
'yellow': color_counter['YELLOW']
})
elif color_counter['RED'] > 0:
raise health_check_base.RedHealthError(
_("Ambari Monitor has responded that cluster has "
"%(red)d critical alert(s)") % {'red': color_counter['RED']})
elif color_counter['YELLOW'] > 0:
raise health_check_base.YellowHealthError(
_("Ambari Monitor has responded that cluster "
"has %d warning alert(s)") % color_counter['YELLOW'])
return _("No alerts found")
def check_health(self):
important_services = self.provider.get_important_services()
observed_data = self.provider.get_health_status(self.service)
imp_map = {'BAD': 'red', 'CONCERNING': 'yellow', 'GOOD': 'green'}
summary = observed_data['summary']
checks = observed_data.get('checks', [])
failed_checks = []
for check in checks:
if check['summary'] != 'GOOD':
failed_checks.append('%(name)s - %(summary)s state' % {
'name': check['name'], 'summary': check['summary']
})
additional_info = None
if failed_checks:
additional_info = _(
"The following checks did not pass: %s") % ",".join(
failed_checks)
if self.service in important_services:
overall = imp_map.get(summary, 'red')
else:
overall = 'green'
if summary != 'GOOD':
overall = 'yellow'
msg = _("Cloudera Manager has responded that service is in "
"the %s state") % summary
if additional_info:
msg = _("%(problem)s. %(description)s") % {
'problem': msg, 'description': additional_info}
if overall == 'red':
raise health_check_base.RedHealthError(msg)
elif overall == 'yellow':
raise health_check_base.YellowHealthError(msg)
return msg
def check_health(self):
instances = self.cluster_context.get_instances(
node_process=self.process)
active_count = 0
for instance in instances:
status = self.process.status(instance)
if status == np.Status.RUNNING:
active_count += 1
if active_count == 0:
if self.process.ui_name in self.IMPORTANT_PROCESSES:
raise health_check_base.RedHealthError(_(
"%s is not in running state") % self.process.ui_name)
else:
raise health_check_base.YellowHealthError(_(
"%s is not in running state") % self.process.ui_name)
if active_count < len(instances):
if self.process.ui_name in self.IMPORTANT_PROCESSES:
raise health_check_base.YellowHealthError(_(
"Some %s processes are not in running state")
% self.process.ui_name)
return _("%s is in running state") % self.process.ui_name
def check_health(self):
instances = self.cluster_context.get_instances(
node_process=management.ZOOKEEPER)
active_count = 0
for instance in instances:
if self._is_zookeeper_running(instance):
active_count += 1
if active_count == 0:
raise health_check_base.RedHealthError(_(
"ZooKeeper is not in running state"))
if active_count < len(instances):
raise health_check_base.YellowHealthError(_(
"Some ZooKeeper processes are not in running state"))
return _("ZooKeeper is in running state")
def check_health(self):
raise health_check_base.YellowHealthError("No problems, boss!")
