from __future__ import unicode_literals import sys from datetime import datetime, timedelta CHECK_INTERVAL = 10 # minutes; The actual interval are 5 minutes but as other # tests might restart Icinga we need to take any # rescheduling into account TABLE_PREFIX = 'icinga_' TABLES = [ # Central tables 'instances', 'objects', # Debugging tables 'conninfo', # Historical tables 'acknowledgements', 'commenthistory', 'contactnotifications', 'dbversion', 'downtimehistory', 'eventhandlers', 'externalcommands', 'flappinghistory', 'hostchecks', 'logentries', 'notifications', 'processevents', 'servicechecks', 'statehistory', 'systemcommands', # Current status tables 'comments', 'customvariablestatus', 'hoststatus', 'programstatus', 'runtimevariables', 'scheduleddowntime', 'servicestatus', 'contactstatus', # Configuration tables 'commands', 'configfiles', 'configfilevariables', 'contact_addresses', 'contact_notificationcommands', 'contactgroup_members', 'contactgroups', 'contactnotificationmethods', 'contacts', 'customvariables', 'host_contactgroups', 'host_contacts', 'host_parenthosts', 'hostdependencies', 'hostescalation_contactgroups', 'hostescalation_contacts', 'hostescalations', 'hostgroup_members', 'hostgroups', 'hosts', 'service_contactgroups', 'service_contacts', 'servicedependencies', 'serviceescalation_contactgroups', 'serviceescalation_contacts', 'serviceescalations', 'servicegroup_members', 'servicegroups', 'services', 'timeperiod_timeranges', 'timeperiods' ] EXAMPLE_CONFIG = { 'localhost': ['disk', 'http', 'icinga', 'load', 'ping4', 'ping6', 'processes', 'ssh', 'users'], 'nsca-ng': ['PassiveService1', 'PassiveService2'] } def success(msg): print '[OK] {0}'.format(msg).encode('utf-8') sys.stdout.flush() return True def fail(msg): print '[FAIL] {0}'.format(msg).encode('utf-8') sys.stdout.flush() return False def info(msg): print '[INFO] {0}'.format(msg).encode('utf-8') sys.stdout.flush() def validate_tables(tables): """ Return whether all tables of the IDO database scheme exist in the given table listing """ info('Checking database scheme... (tables)') failures = False for table in (TABLE_PREFIX + n for n in TABLES): if table in tables: success('Found table "{0}" in database'.format(table)) else: fail('Could not find table "{0}" in database'.format(table)) failures = True return not failures def verify_host_config(config_data): """ Return whether the example hosts exist in the given "hosts" table """ info('Checking example host configuration...') failures = False for hostname in EXAMPLE_CONFIG: if not any(1 for e in config_data if e['alias'] == hostname): fail('Could not find host "{0}"'.format(hostname)) failures = True else: success('Found host "{0}"'.format(hostname)) return not failures def verify_service_config(config_data): """ Return whether the example services exist in the given "services" table """ info('Checking example service configuration...') failures = False for hostname, servicename in ((h, s) for h, ss in EXAMPLE_CONFIG.iteritems() for s in ss): if not any(1 for c in config_data if c['alias'] == hostname and c['display_name'] == servicename): fail('Could not find service "{0}" on host "{1}"' ''.format(servicename, hostname)) failures = True else: success('Found service "{0}" on host "{1}"' ''.format(servicename, hostname)) return not failures def check_last_host_status_update(check_info): """ Return whether the example hosts are checked as scheduled """ info('Checking last host status updates...') failures = False for host_info in check_info: if host_info['alias'] == 'localhost': last_check = datetime.fromtimestamp(float(host_info['last_check'])) if datetime.now() - last_check > timedelta(minutes=CHECK_INTERVAL, seconds=10): fail('The last status update of host "{0}" was more than {1} ' 'minutes ago'.format(host_info['alias'], CHECK_INTERVAL)) failures = True else: success('Host "{0}" is being updated'.format(host_info['alias'])) elif host_info['alias'] == 'nsca-ng': if float(host_info['last_check']) > 0: fail('The host "{0}" was checked even though it has' ' no check service'.format(host_info['alias'])) failures = True else: success('Host "{0}" is not being checked because there ' 'is no check service'.format(host_info['alias'])) else: info('Skipping host "{0}"'.format(host_info['alias'])) return not failures def check_last_service_status_update(check_info): """ Return whether the example services are checked as scheduled """ info('Checking last service status updates...') failures = False for svc_info in check_info: if svc_info['display_name'] in EXAMPLE_CONFIG.get(svc_info['alias'], []): last_check = datetime.fromtimestamp(float(svc_info['last_check'])) if datetime.now() - last_check > timedelta(minutes=CHECK_INTERVAL, seconds=10): fail('The last status update of service "{0}" on' ' host "{1}" was more than {2} minutes ago' ''.format(svc_info['display_name'], svc_info['alias'], CHECK_INTERVAL)) failures = True else: success('Service "{0}" on host "{1}" is being updated' ''.format(svc_info['display_name'], svc_info['alias'])) else: info('Skipping service "{0}" on host "{1}"' ''.format(svc_info['display_name'], svc_info['alias'])) return not failures def check_logentries(logentry_info): """ Return whether the given logentry originates from host "localhost" and refers to its very last hard status change """ info('Checking status log for host "localhost"...') if logentry_info and logentry_info[0]['alias'] == 'localhost': entry_time = datetime.fromtimestamp(float(logentry_info[0]['entry_time'])) state_time = datetime.fromtimestamp(float(logentry_info[0]['state_time'])) if entry_time - state_time > timedelta(seconds=10): return fail('The last hard state of host "localhost"' ' seems not to have been logged') else: return fail('No logs found in the IDO for host "localhost"') return success('The last hard state of host "localhost" was properly logged')