def schedule_servicegroup_service_downtime(
connection,
servicegroup_name: str,
start_time: dt.datetime,
end_time: dt.datetime,
include_hosts: bool = False,
recur: RecurMode = 'fixed',
trigger_id: int = 0,
duration: int = 0,
user_id: str = '',
comment: str = '',
):
"""Schedules downtime for all hosts, which have services in a given servicegroup.
Args:
connection:
A LiveStatus connection object.
servicegroup_name:
The name of the service group. Any host having a service in this group will be
A downtime will be scheduled for all hosts in this group.
start_time:
When the downtime shall begin.
end_time:
When the downtime shall end.
include_hosts:
When set to True, all hosts will also receive a scheduled downtime, not just their
services which belong to this service group.
recur:
The recurring mode of the new downtime. Available modes are:
* fixed
* hour
* day
* week
* second_week
* fourth_week
* weekday_start
* weekday_end
* day_of_month
This only works when using the Enterprise Editions. Defaults to 'fixed'.
trigger_id:
The id of another downtime-entry. If given (other than 0) then this downtime will be
triggered by the other downtime.
duration:
Duration in seconds. When set, the downtime does not begin automatically at a nominated
time, but when a real problem status appears for the host. Consequently, the
start_time/end_time is only the time window in which the scheduled downtime can begin.
user_id:
comment:
A comment which will be added to the downtime.
connection:
"""
members: List[List[str]] = Query(
[tables.Servicegroups.members],
tables.Servicegroups.name.equals(servicegroup_name),
).value(connection)
for host_name, service_description in members:
schedule_service_downtime(
connection,
host_name=host_name,
service_description=service_description,
start_time=start_time,
end_time=end_time,
recur=recur,
trigger_id=trigger_id,
duration=duration,
user_id=user_id,
comment=comment,
)
if include_hosts:
host_names = _deduplicate([_host_name for _host_name, _ in members])
schedule_host_downtime(
connection,
host_name=host_names,
start_time=start_time,
end_time=end_time,
recur=recur,
trigger_id=trigger_id,
duration=duration,
user_id=user_id,
comment=comment,
)
def schedule_hostgroup_host_downtime(
connection,
hostgroup_name: str,
start_time: dt.datetime,
end_time: dt.datetime,
include_all_services: bool = False,
recur: RecurMode = 'fixed',
trigger_id: int = 0,
duration: int = 0,
user_id: str = '',
comment: str = '',
):
"""Schedules downtime for all hosts in a given hostgroup.
Args:
connection:
A LiveStatus connection object.
hostgroup_name:
The name of the hostgroup. A downtime will be scheduled for all hosts in this hostgroup.
start_time:
When the downtime shall begin.
end_time:
When the downtime shall end.
include_all_services:
If set, downtimes for all services associated with the given host will be scheduled.
Defaults to False.
recur:
The recurring mode of the new downtime. Available modes are:
* fixed
* hour
* day
* week
* second_week
* fourth_week
* weekday_start
* weekday_end
* day_of_month
This only works when using the Enterprise Editions. Defaults to 'fixed'.
trigger_id:
The id of another downtime-entry. If given (other than 0) then this downtime will be
triggered by the other downtime.
duration:
Duration in seconds. When set, the downtime does not begin automatically at a nominated
time, but when a real problem status appears for the host. Consequently, the
start_time/end_time is only the time window in which the scheduled downtime can begin.
user_id:
comment:
A comment which will be added to the downtime.
connection:
See Also:
* https://assets.nagios.com/downloads/nagioscore/docs/externalcmds/cmdinfo.php?command_id=123
"""
members: List[str] = Query([tables.Hostgroups.members],
tables.Hostgroups.name.equals(hostgroup_name)).value(connection)
schedule_host_downtime(
connection,
host_name=members,
start_time=start_time,
end_time=end_time,
include_all_services=include_all_services,
recur=recur,
trigger_id=trigger_id,
duration=duration,
user_id=user_id,
comment=comment,
)
def acknowledge_servicegroup_problem(
connection,
servicegroup_name: str,
sticky: bool = False,
notify: bool = False,
persistent: bool = False,
user: str = "",
comment: str = "",
):
"""Acknowledge the problems of the current services of the service group
When acknowledging a problem, further notifications for the respective services are disabled, as
long as a specific service doesn't change state. At state change, notifications are re-enabled.
Args:
connection:
A livestatus connection object.
servicegroup_name:
The host-name for which this acknowledgement is for.
sticky:
If set, only a state-change of the service to an OK state will discard the
acknowledgement. Otherwise it will be discarded on any state-change. Defaults to False.
notify:
If set, notifications will be sent out to the configured contacts. Defaults to False.
persistent:
If set, the comment will persist a restart. Defaults to False.
user:
comment:
If set, this comment will be stored alongside the acknowledgement.
Raises:
ValueError:
When the service group could not be found.
"""
with detailed_connection(connection) as conn:
group_entries = Query(
[tables.Servicegroups.members],
tables.Servicegroups.name.equals(servicegroup_name),
).fetchall(conn)
acknowledgement = 2 if sticky else 1 # 1: normal, 2: sticky
for entry in group_entries:
site_id = entry["site"]
for host_name, service_description in entry["members"]:
send_command(
connection,
"ACKNOWLEDGE_SVC_PROBLEM",
[
host_name,
service_description,
acknowledgement,
int(notify),
int(persistent),
user,
comment,
],
site_id=site_id,
)
def host_is_monitored(host_name: str) -> bool:
return bool(
Query([Hosts.name], Hosts.name == host_name).first_value(sites.live()))
def acknowledge_host_problem(
connection,
host_name,
sticky: bool = False,
notify: bool = False,
persistent: bool = False,
user: str = "",
comment: str = "",
):
"""Acknowledge the current problem for the given host.
When acknowledging a problem, notifications for the host are disabled, as long as the
host doesn't change state. At state change, notifications are re-enabled.
Args:
connection:
A livestatus connection object.
host_name:
The host-name for which this acknowledgement is for.
sticky:
If set, only a state-change of the host to an UP state will discard the acknowledgement.
Otherwise it will be discarded on any state-change. Defaults to False.
notify:
If set, notifications will be sent out to the configured contacts. Defaults to False.
persistent:
If set, the comment will persist a restart. Defaults to False.
user:
comment:
If set, this comment will be stored alongside the acknowledgement.
Examples:
>>> from cmk.gui.livestatus_utils.testing import simple_expect
>>> cmd = "COMMAND [...] ACKNOWLEDGE_HOST_PROBLEM;example.com;1;0;0;;"
>>> with simple_expect() as live:
... _ = live.expect_query("GET hosts\\nColumns: name\\nFilter: name = example.com")
... _ = live.expect_query(cmd, match_type="ellipsis")
... acknowledge_host_problem(live, 'example.com')
"""
acknowledgement = 2 if sticky else 1 # 1: normal, 2: sticky
with detailed_connection(connection) as conn:
site_id = Query([Hosts.name], Hosts.name.equals(host_name)).first_value(conn)
return send_command(
connection,
"ACKNOWLEDGE_HOST_PROBLEM",
[
host_name,
acknowledgement,
int(notify),
int(persistent),
user,
comment,
],
site_id=site_id,
)
def _query_site(connection, host_name: str) -> str:
with detailed_connection(connection) as conn:
site_id = Query([Hosts.name], Hosts.name.equals(host_name)).first_value(conn)
if not isinstance(site_id, str):
raise QueryException
return site_id
def create_service_related_downtime(params):
"""Create a service related scheduled downtime"""
body = params["body"]
live = sites.live()
downtime_type: DowntimeType = body["downtime_type"]
if downtime_type == "service":
host_name = body["host_name"]
with detailed_connection(live) as conn:
site_id = Query(columns=[Hosts.name], filter_expr=Hosts.name.op("=", host_name)).value(
conn
)
downtime_commands.schedule_service_downtime(
live,
site_id,
host_name=body["host_name"],
service_description=body["service_descriptions"],
start_time=body["start_time"],
end_time=body["end_time"],
recur=body["recur"],
duration=body["duration"],
user_id=user.ident,
comment=body.get(
"comment",
f"Downtime for services {', '.join(body['service_descriptions'])!r}@{body['host_name']!r}",
),
)
elif downtime_type == "servicegroup":
downtime_commands.schedule_servicegroup_service_downtime(
live,
servicegroup_name=body["servicegroup_name"],
start_time=body["start_time"],
end_time=body["end_time"],
recur=body["recur"],
duration=body["duration"],
user_id=user.ident,
comment=body.get("comment", f"Downtime for servicegroup {body['servicegroup_name']!r}"),
)
elif downtime_type == "service_by_query":
try:
downtime_commands.schedule_services_downtimes_with_query(
live,
query=body["query"],
start_time=body["start_time"],
end_time=body["end_time"],
recur=body["recur"],
duration=body["duration"],
user_id=user.ident,
comment=body.get("comment", ""),
)
except QueryException:
return problem(
status=422,
title="Query did not match any service",
detail="The provided query returned an empty list so no downtime was set",
)
else:
return problem(
status=400,
title="Unhandled downtime-type.",
detail=f"The downtime-type {downtime_type!r} is not supported.",
)
return Response(status=204)
def set_acknowledgement_on_hosts(params):
"""Set acknowledgement on related hosts"""
body = params["body"]
live = sites.live()
sticky = body["sticky"]
notify = body["notify"]
persistent = body["persistent"]
comment = body["comment"]
acknowledge_type = body["acknowledge_type"]
if acknowledge_type == "host":
name = body["host_name"]
host_state = Query([Hosts.state], Hosts.name == name).value(live)
if not host_state:
raise ProblemException(
status=422,
title=f"Host {name!r} has no problem.",
)
acknowledge_host_problem(
live,
name,
sticky=sticky,
notify=notify,
persistent=persistent,
user=user.ident,
comment=comment,
)
elif acknowledge_type == "hostgroup":
host_group = body["hostgroup_name"]
try:
acknowledge_hostgroup_problem(
live,
host_group,
sticky=sticky,
notify=notify,
persistent=persistent,
user=user.ident,
comment=comment,
)
except ValueError:
raise ProblemException(
400,
title="Host group could not be found.",
detail=f"Unknown host group: {host_group}",
)
elif acknowledge_type == "host_by_query":
query = body["query"]
hosts = Query([Hosts.name], query).fetchall(live)
if not hosts:
raise ProblemException(
status=422,
title="The provided query returned no monitored hosts",
)
for host in hosts:
acknowledge_host_problem(
live,
host.name,
sticky=sticky,
notify=notify,
persistent=persistent,
user=user.ident,
comment=comment,
)
else:
raise ProblemException(
status=400,
title="Unhandled acknowledge-type.",
detail=
f"The acknowledge-type {acknowledge_type!r} is not supported.",
)
return http.Response(status=204)
def set_acknowledgement_on_services(params):
"""Set acknowledgement on related services"""
body = params["body"]
live = sites.live()
sticky = body["sticky"]
notify = body["notify"]
persistent = body["persistent"]
comment = body["comment"]
acknowledge_type = body["acknowledge_type"]
if acknowledge_type == "service":
description = unquote(body["service_description"])
host_name = body["host_name"]
service = Query(
[Services.host_name, Services.description, Services.state],
And(Services.host_name == host_name,
Services.description == description),
).first(live)
if not service:
raise ProblemException(
status=400,
title=
f"Service {description!r}@{host_name!r} could not be found.",
)
if not service.state:
raise ProblemException(
status=422,
title=f"Service {description!r}@{host_name!r} has no problem.",
)
acknowledge_service_problem(
live,
service.host_name,
service.description,
sticky=sticky,
notify=notify,
persistent=persistent,
user=user.ident,
comment=comment,
)
elif acknowledge_type == "servicegroup":
service_group = body["servicegroup_name"]
try:
acknowledge_servicegroup_problem(
live,
service_group,
sticky=sticky,
notify=notify,
persistent=persistent,
user=user.ident,
comment=comment,
)
except ValueError:
raise ProblemException(
status=400,
title="Service group could not be found.",
detail=f"Unknown service group: {service_group}",
)
elif acknowledge_type == "service_by_query":
services = Query(
[Services.host_name, Services.description, Services.state],
body["query"],
).fetchall(live)
if not services:
raise ProblemException(
status=422,
title="No services with problems found.",
detail="All queried services are OK.",
)
for service in services:
if not service.state:
continue
acknowledge_service_problem(
live,
service.host_name,
service.description,
sticky=sticky,
notify=notify,
persistent=persistent,
user=user.ident,
comment=comment,
)
else:
raise ProblemException(
status=400,
title="Unhandled acknowledge-type.",
detail=
f"The acknowledge-type {acknowledge_type!r} is not supported.",
)
return http.Response(status=204)
def set_acknowledgement_on_services(params):
"""Set acknowledgement on related services"""
body = params['body']
live = sites.live()
sticky = body['sticky']
notify = body['notify']
persistent = body['persistent']
comment = body['comment']
acknowledge_type = body['acknowledge_type']
if acknowledge_type == 'service':
description = unquote(body['service_description'])
host_name = body['host_name']
service = Query(
[Services.host_name, Services.description, Services.state],
And(Services.host_name == host_name,
Services.description == description)).first(live)
if not service:
raise ProblemException(
status=400,
title=
f'Service {description!r}@{host_name!r} could not be found.',
)
if not service.state:
raise ProblemException(
status=422,
title=f'Service {description!r}@{host_name!r} has no problem.',
)
acknowledge_service_problem(
live,
service.host_name,
service.description,
sticky=sticky,
notify=notify,
persistent=persistent,
user=config.user.ident,
comment=comment,
)
elif acknowledge_type == 'servicegroup':
service_group = body['servicegroup_name']
try:
acknowledge_servicegroup_problem(
live,
service_group,
sticky=sticky,
notify=notify,
persistent=persistent,
user=config.user.ident,
comment=comment,
)
except ValueError:
raise ProblemException(
status=400,
title="Servicegroup could not be found.",
detail=f"Unknown servicegroup: {service_group}",
)
elif acknowledge_type == 'service_by_query':
services = Query(
[Services.host_name, Services.description, Services.state],
body['query'],
).fetchall(live)
if not services:
raise ProblemException(
status=422,
title='No services with problems found.',
detail='All queried services are OK.',
)
for service in services:
if not service.state:
continue
acknowledge_service_problem(
live,
service.host_name,
service.description,
sticky=sticky,
notify=notify,
persistent=persistent,
user=config.user.ident,
comment=comment,
)
else:
raise ProblemException(
status=400,
title="Unhandled acknowledge-type.",
detail=
f"The acknowledge-type {acknowledge_type!r} is not supported.",
)
return http.Response(status=204)
def schedule_host_downtime(
connection,
host_entry: Union[str, List[str]],
start_time: dt.datetime,
end_time: dt.datetime,
include_all_services: bool = False,
recur: RecurMode = "fixed",
trigger_id: int = 0,
duration: int = 0,
user_id: str = "",
comment: str = "",
):
"""Schedule the downtime of a host.
Notes:
If `include_all_services` is set to True, the services table is only queried
once, instead of len(host_name) times. If a lot of hosts are to be scheduled, this
will save N queries. Issuing the command is still done sequentially.
Args:
connection:
A livestatus connection object.
host_entry:
The host-name for which this downtime is for.
start_time:
When the downtime shall begin.
end_time:
When the downtime shall end.
include_all_services:
If set, downtimes for all services associated with the given host will be scheduled.
Defaults to False.
recur:
The recurring mode of the new downtime. Available modes are:
* fixed
* hour
* day
* week
* second_week
* fourth_week
* weekday_start
* weekday_end
* day_of_month
This only works when using the Enterprise Editions. Defaults to 'fixed'.
trigger_id:
The id of another downtime-entry. If given (other than 0) then this downtime will be
triggered by the other downtime.
duration:
Duration in seconds. When set, the downtime does not begin automatically at a nominated
time, but when a real problem status appears for the host. Consequencely, the
start_time/end_time is only the time window in which the scheduled downtime can begin.
user_id:
comment:
A comment which will be added to the downtime.
See Also:
* https://assets.nagios.com/downloads/nagioscore/docs/externalcmds/cmdinfo.php?command_id=118
* https://assets.nagios.com/downloads/nagioscore/docs/externalcmds/cmdinfo.php?command_id=122
Examples:
>>> import pytz
>>> _start_time = dt.datetime(1970, 1, 1, tzinfo=pytz.timezone("UTC"))
>>> _end_time = dt.datetime(1970, 1, 2, tzinfo=pytz.timezone("UTC"))
>>> from cmk.gui.livestatus_utils.testing import simple_expect
>>> from cmk.gui.config import load_config
>>> from cmk.gui.utils.script_helpers import application_and_request_context
>>> from cmk.gui.logged_in import SuperUserContext
>>> cmd = "COMMAND [...] SCHEDULE_HOST_DOWNTIME;example.com;0;86400;16;0;120;;Boom"
>>> with simple_expect() as live, application_and_request_context(), SuperUserContext():
... load_config()
... _ = live.expect_query("GET hosts\\nColumns: name\\nFilter: name = example.com")
... _ = live.expect_query(cmd, match_type="ellipsis")
... schedule_host_downtime(live,
... 'example.com',
... _start_time,
... _end_time,
... recur="day_of_month",
... duration=120,
... comment="Boom")
"""
if isinstance(host_entry, str):
hosts = [host_entry]
elif host_entry:
hosts = host_entry
else:
raise ValueError("List of hosts may not be empty.")
with detailed_connection(connection) as conn:
host_entries = [(entry["site"], entry["name"]) for entry in Query(
[Hosts.name], Or(*[Hosts.name.equals(host)
for host in hosts])).fetchall(conn)]
for _site, _host_name in host_entries:
_schedule_downtime(
connection,
"SCHEDULE_HOST_DOWNTIME",
_site,
_host_name,
None,
start_time,
end_time,
recur,
trigger_id,
duration,
user_id,
comment,
)
if include_all_services:
with detailed_connection(connection) as conn:
services = Query(
[tables.Services.host_name, tables.Services.description],
Or(*[
tables.Services.host_name.equals(_host_name)
for _, _host_name in host_entries
]),
).fetch_values(conn)
for _site, _host_name, service_description in services:
schedule_service_downtime(
connection,
_site,
host_name=_host_name,
service_description=service_description,
start_time=start_time,
end_time=end_time,
recur=recur,
trigger_id=trigger_id,
duration=duration,
user_id=user_id,
comment=comment,
)