Skip to main content
PUT
Update a monitor

Authorizations

​
X-BRONTO-API-KEY
string
header
required

Path Parameters

​
monitorId
string<uuid>
required

The ID of the monitor to update

Body

application/json
​
name
string
required

The name of the monitor

Required string length: 1 - 255
Example:

"High Error Rate"

​
comparison_operator
enum<string>
required

the comparison operator of the monitor (ABOVE, ABOVE_OR_EQUAL, BELOW, BELOW_OR_EQUAL, EQUAL, NOT_EQUAL, OUTSIDE). OUTSIDE is two-sided and is only allowed for ANOMALY_DETECTION monitors, which are the only type whose compared value is a signed deviation from a baseline.

Available options:
BELOW,
BELOW_OR_EQUAL,
ABOVE,
ABOVE_OR_EQUAL,
EQUAL,
NOT_EQUAL,
OUTSIDE
Example:

"ABOVE"

​
threshold
number
required

the threshold the monitor compares against. For PATTERN, USAGE and CHANGE_DETECTION monitors this is a metric value. For ANOMALY_DETECTION monitors it is instead a count of baseline spread units (scaled median absolute deviations, comparable to standard deviations), so it must be greater than 0; a threshold of 3 means "more than 3 spread units away from the baseline centre".

Example:

1000

​
window
string
required

The time window to evaluate the monitor; minimum five minutes, maximum one day

Example:

"'Last 20 minutes', 'Last 2 hours', 'Last 3 days'"

​
actions
object[]
required

The list of actions (notifications) to take when the monitor fires or resolves

Required array length: 1 - 10 elements
​
queries
object[]
required

List of queries to use for monitor

Minimum array length: 1
​
description
string

Extra contextual information to add to the monitor.

Required string length: 1 - 4000
Example:

"Average response time is over 1 seconds in the last 20 minutes"

​
warning_threshold
number

the warning threshold value, in the same units as threshold. For ANOMALY_DETECTION monitors it must be greater than 0 and less than threshold whatever the comparison operator, because the comparison is on the magnitude of the deviation.

Example:

1000

​
no_data_status
enum<string>

Specify the status to use in case of no data (no events matched in a time window)

Available options:
OK,
NO_DATA,
ALERT,
WARN
Example:

"OK"

​
notify_once
boolean
default:false

Notify only the groups changing the global monitor status, i.e. notifications are sent only for the first group entering ALERT status or for the last group resolving

Example:

true

​
group_retention
integer<int64>

Remove groups with no data after the provided time in milliseconds

Required range: 0 <= x <= 604800000
Example:

600000

​
aux
object

Additional monitor configuration

​
monitor_type
enum<string>
default:PATTERN

What the monitor compares its metric against. PATTERN and USAGE compare it to a fixed threshold, CHANGE_DETECTION to the same metric at an earlier time (see compare_to and change_type), and ANOMALY_DETECTION to a trailing baseline of the same series (see evaluation_window and anomaly_algorithm).

Available options:
PATTERN,
USAGE,
CHANGE_DETECTION,
ANOMALY_DETECTION
Example:

"PATTERN"

​
compare_to
string

For change detection monitors only. How far back in time the comparison query should be run. Minimum "5 mins ago", maximum "1 month".

Example:

"'2 hours ago', '1 days ago'"

​
change_type
enum<string>

For change detection monitors only. Specifies how the change values are calculated, as described in the Bronto documentation.

Available options:
DIFFERENCE,
PERCENTAGE
​
evaluation_window
string

For anomaly detection monitors only, where it is required. The trailing period used to build the baseline the candidate window is compared against, expressed as " ". It must cover at least 10 and at most 1000 monitor windows and be at most 30 days: fewer than 10 windows leaves the baseline permanently degenerate, so the monitor reports healthy and never fires.

Example:

"'24 hours', '7 days'"

​
anomaly_algorithm
enum<string>
default:MEDIAN_MAD

For anomaly detection monitors only. How the baseline centre and spread are estimated. MEDIAN_MAD uses the median as the centre and the median absolute deviation, scaled by 1.4826 so it reads like a standard deviation, as the spread.

Available options:
MEDIAN_MAD
Example:

"MEDIAN_MAD"

​
ai_report_enabled
boolean

True if an automatic report should be generated when the monitor triggers

​
ai_report_instructions
string

Instructions for the AI on what to include in the report, e.g. check for recent deployments, verify the health of the following dependencies: ... etc

​
formulas
object[]

List of formulas to use for monitor

Response

Monitor is successfully updated

​
id
string<uuid>
required

The unique identifier for the monitor

​
name
string
required

The name of the monitor

Required string length: 1 - 255
Example:

"High Error Rate"

​
comparison_operator
enum<string>
required

the comparison operator of the monitor (ABOVE, ABOVE_OR_EQUAL, BELOW, BELOW_OR_EQUAL, EQUAL, NOT_EQUAL, OUTSIDE). OUTSIDE is two-sided and is only allowed for ANOMALY_DETECTION monitors, which are the only type whose compared value is a signed deviation from a baseline.

Available options:
BELOW,
BELOW_OR_EQUAL,
ABOVE,
ABOVE_OR_EQUAL,
EQUAL,
NOT_EQUAL,
OUTSIDE
Example:

"ABOVE"

​
threshold
number<double>
required

the threshold the monitor compares against. For PATTERN, USAGE and CHANGE_DETECTION monitors this is a metric value. For ANOMALY_DETECTION monitors it is instead a count of baseline spread units (scaled median absolute deviations, comparable to standard deviations), so it must be greater than 0; a threshold of 3 means "more than 3 spread units away from the baseline centre".

Example:

1000

​
window
string
required

The time window to evaluate the monitor; minimum five minutes, maximum one day

Example:

"'Last 20 minutes', 'Last 2 hours', 'Last 1 days'"

​
actions
object[]
required

The list of actions (notifications) to take when the monitor fires or resolves

Required array length: 1 - 10 elements
​
status
enum<string>
required

The status of the monitor

Available options:
OK,
NO_DATA,
ALERT,
WARN
Example:

"OK"

​
monitor_type
enum<string>
default:PATTERN
required

What the monitor compares its metric against. PATTERN and USAGE compare it to a fixed threshold, CHANGE_DETECTION to the same metric at an earlier time (see compare_to and change_type), and ANOMALY_DETECTION to a trailing baseline of the same series (see evaluation_window and anomaly_algorithm).

Available options:
PATTERN,
USAGE,
CHANGE_DETECTION,
ANOMALY_DETECTION
Example:

"PATTERN"

​
ai_report_enabled
boolean
required

True if an automatic report should be generated when the monitor triggers

​
metadata
object
required

Common metadata attached to all persisted resources.

​
description
string

Extra contextual information to add to the monitor

Maximum string length: 4000
Example:

"Average response time is over 1 seconds in the last 20 minutes"

​
warning_threshold
number<double>

the warning threshold value, in the same units as threshold. For ANOMALY_DETECTION monitors it must be greater than 0 and less than threshold whatever the comparison operator, because the comparison is on the magnitude of the deviation.

Example:

900

​
monitored_groups
object[]

Representation of all the monitored groups and their status

Example:
​
no_data_status
enum<string>

Specify the status to use if a group has no data (no events matched in a time window)

Available options:
OK,
NO_DATA,
ALERT,
WARN
Example:

"OK"

​
muted_until
integer<int64>

The timestamp indicating the ending of the muting. -1 indicates muted forever

Example:

1710958395538

​
notify_once
boolean
default:false

Notify only the groups changing the global monitor status, i.e. notifications are sent only for the first group entering ALERT status or for the last group resolving

Example:

false

​
group_retention
integer<int64>

Remove groups with no data after the provided time in milliseconds

Required range: 0 <= x <= 604800000
Example:

600000

​
last_trigger_ts
integer<int64>

The timestamp of the last time the monitor fired

Example:

1710948395538

​
aux
object

Additional monitor configuration

​
compare_to
string

For change detection monitors only. How far back in time the comparison query should be run. Minimum "5 mins ago", maximum "1 month".

Example:

"2 hours ago"

​
change_type
enum<string>

For change detection monitors only. Specifies how the change values are calculated, as described in the Bronto documentation.

Available options:
DIFFERENCE,
PERCENTAGE
​
evaluation_window
string

For anomaly detection monitors only, where it is required. The trailing period used to build the baseline the candidate window is compared against, expressed as " ". It must cover at least 10 and at most 1000 monitor windows and be at most 30 days: fewer than 10 windows leaves the baseline permanently degenerate, so the monitor reports healthy and never fires.

Example:

"'24 hours', '7 days'"

​
anomaly_algorithm
enum<string>
default:MEDIAN_MAD

For anomaly detection monitors only. How the baseline centre and spread are estimated. MEDIAN_MAD uses the median as the centre and the median absolute deviation, scaled by 1.4826 so it reads like a standard deviation, as the spread.

Available options:
MEDIAN_MAD
Example:

"MEDIAN_MAD"

​
ai_report_instructions
string

Instructions for the AI on what to include in the report, e.g. check for recent deployments, verify the health of the following dependencies: ... etc

​
metric_id
string<uuid>

The unique identifier for the metric definition that backs this monitor

​
template
object