Skip to main content

Create an alert

POST 
/v2/alerts

Configures and activates a new alert. Please ensure to change the region in the URL to match your account's region.

Request​

Body

    title stringrequired

    Alert title

    description string

    A description of the event, its significance, and suggested next steps or instructions for the team.

    tags string[]

    Possible values: <= 10

    Tags for filtering alerts and triggered alerts. Can be used in Kibana Discover, dashboards, and more.

    output object

    Automatically sends out notifications with sample results when the alert triggers.

    recipients object

    Add email addresses and/or endpoint channels to automatically receive notifications with sample data when the alert triggers.

    emails string[]

    Array of email addresses to be notified when the alert triggers.

    notificationEndpointIds int32[]

    Array of IDs of pre-configured endpoint channels to notify when the alert triggers.

    suppressNotificationsMinutes int32

    Possible values: >= 5 and <= 1440

    Add a waiting period in minutes to space out notifications. (The alert will still trigger but will not send out notifications during the waiting period.)

    type string

    Possible values: [JSON, TABLE]

    Selects the output format for the alert notification. If the alert has no aggregations/group by fields, JSON offers the option to send full sample logs without selecting specific fields.

    searchTimeFrameMinutes int32

    Possible values: >= 5 and <= 1440

    The time frame for evaluating the log data is a sliding window, with 1 minute granularity.

    The recommended minimum and maximum values are not validated, but needed to guarantee the alert's accuracy.

    The minimum recommended time frame is 5 minutes, as anything shorter will be less reliable and unnecessarily resource-heavy.

    The maximum recommended time frame is 1440 minutes (24 hours). The alert runs on the index from today and yesterday (in UTC) and the maximum time frame increases throughout the day, reaching 48 hours exactly before midnight UTC.

    The default value is 5.

    subComponents object[]required

    Sets the search criteria using a search query, filters, group by aggregations, accounts to search, and trigger conditions.

  • Array [
    • queryDefinition object

    Determines when the alert should trigger using any combination of a search query, filters, group by aggregations, accounts to search, and trigger conditions.

    query string

    Default value: *

    Provide a Kibana search query written in Lucene syntax. The search query together with the filters select for the relevant logs.

    Cannot be null - send an asterisk wildcard * if not using a search query.

    filters object

    Apply must and must_not filters to the monitoring alert. Filters are more efficient compared to a query, so it's recommended to opt for a filter over a query, where possible. See Elasticsearch Bool-Query for more detail.

    bool object

    Runs Elasticsearch Bool Query filters on the data (before the search query is applied). The most efficient way to grab the logs you are looking for.

    must object[]
  • Array [
    • match_phrase object
    Field object
    query string
  • ]
    • must_not object[]
  • Array [
    • match_phrase object
    Field object
    query string
  • ]
    • groupBy string[]nullable

    Possible values: <= 3

    Specify 1-3 fields by which to group the results and count them. If you apply a group by operation, the alert returns a count of the results aggregated by unique values.

    aggregation object

    Specifies a trigger condition that acts as a threshold.

    aggregationType string

    Possible values: [SUM, MIN, MAX, AVG, COUNT, UNIQUE_COUNT, NONE, PERCENTAGE, PERCENTILE]

    Specifies the aggregation operator.

    • If COUNT, fieldToAggregateOn must be null, and groupBy fields must not be empty.

    • If NONE, fieldToAggregateOn must be null, and groupBy field must not be empty (or null).

    • If PERCENTAGE, valueToAggregateOn must be specified.

    • If any other operator type (other than NONE or COUNT), fieldToAggregateOn must not be null.

    fieldToAggregateOn string

    Selects the field on which to run the aggregation for the trigger condition.

    • Cannot be a field already in use for groupBy.
    valueToAggregateOn string

    Used by the PERCENTAGE aggregation to select the field’s value. This value is used to determine if its ratio out of the total amount of logs in the query satisfies the trigger condition.

    • Only relevant for the PERCENTAGE aggregation.
    shouldQueryOnAllAccounts boolean

    Default value: true

    Only applicable when the alert is run from the main account. If true, the alert runs on the main account and all associated searchable sub accounts. If false, specify relevant account IDs for the alert to monitor using the accountIdsToQueryOn field.

    accountIdsToQueryOn int32[]

    Specify Account IDs to select which accounts the alert should monitor. The alert will be checked only on these accounts.

    trigger object

    Sets the triggering threshold and severity tab to label the event when the alert triggers.

    operator string

    Possible values: [LESS_THAN, GREATER_THAN, LESS_THAN_OR_EQUALS, GREATER_THAN_OR_EQUALS, EQUALS, NOT_EQUALS]

    Specifies the operator for evaluating the results.

    severityThresholdTiers object

    Possible values: [INFO, LOW, MEDIUM, HIGH, SEVERE]

    Default value: [object Object]

    Sets a severity label per trigger threshold as a key:value pair.

    If using more than one sub-component, only 1 severityThresholdTiers is allowed. Otherwise, 1 per enum are allowed (for a total of 5 thresholds of increasing severities).

    Increasing severity must adhere to the logic of the operator.

    output object

    Selects the data output to be sent in the notification when the alert triggers. Not applicable, when grouping by fields or aggregating results, as the output is auto-selected.

    shouldUseAllFields boolean

    Default value: true

    If true, the notification output will include entire logs with all of their fields in the sample data.

  • ]
    • correlations object

    Only applicable when multiple sub-components are in use. Selects a logic for correlating the alert’s sub-components.

    AND is currently the only supported operator. When AND is the correlationOperator, both sub-components must meet their triggering criteria for the alert to trigger.

    correlationOperators string[]

    Possible values: [AND]

    joins object[]

    Specifies which group by fields must have the same values to trigger the alert.

    Joins the group by fields from the first and second sub-components. The key represents the index of the sub component in the array (See the example - the index of the first sub-component is 0, the second is 1).

    The fields must be ordered pairs of the group by fields already in use in the queryDefinition.

    schedule object

    Defines the frequency and the time frame in which an alert will be evaluated.

    cronExpression string

    Cron job for the intervals schedule.

    timezone string

    Time zone for the cron job. If no time zone is selected, UTC will be used by default.

    enabled boolean

    If true, the alert is enabled and active. Default value is true.

Responses​

successful operation

Schema
    id int32

    Logz.io alert ID.

    updatedAt string

    Date and time in UTC when the alert was last updated.

    updatedBy string

    Email of the user who last updated the alert.

    createdAt string

    Date and time in UTC when the alert was first created.

    createdBy string

    Email of the user who first created the alert.

    enabled boolean

    If true, the alert is currently active.

    title string

    Alert title.

    description string

    A description of the event, its significance, and suggested next steps or instructions for the team.

    tags string[]

    Tags for filtering alerts and triggered alerts. Can be used in Kibana Discover, dashboards, and more.

    output object

    Automatically sends out notifications with sample results when the alert triggers.

    recipients object

    Add email addresses and/or endpoint channels to automatically receive notifications with sample data when the alert triggers.

    emails string[]

    Array of email addresses to be notified when the alert triggers.

    notificationEndpointIds int32[]

    Array of IDs of pre-configured endpoint channels to notify when the alert triggers.

    suppressNotificationsMinutes int32

    Possible values: >= 5 and <= 1440

    Add a waiting period in minutes to space out notifications. (The alert will still trigger but will not send out notifications during the waiting period.)

    type string

    Possible values: [JSON, TABLE]

    Selects the output format for the alert notification. If the alert has no aggregations/group by fields, JSON offers the option to send full sample logs without selecting specific fields.

    searchTimeFrameMinutes int32

    Possible values: >= 5 and <= 1440

    The time frame for evaluating the log data is a sliding window, with 1 minute granularity.

    The recommended minimum and maximum values are not validated, but needed to guarantee the alert's accuracy.

    The minimum recommended time frame is 5 minutes, as anything shorter will be less reliable and unnecessarily resource-heavy.

    The maximum recommended time frame is 1440 minutes (24 hours). The alert runs on the index from today and yesterday (in UTC) and the maximum time frame increases throughout the day, reaching 48 hours exactly before midnight UTC.

    subComponents object[]

    Determines when the alert should trigger using any combination of a search query, filters, group by aggregations, accounts to search, and trigger conditions.

  • Array [
    • queryDefinition object

    Determines when the alert should trigger using any combination of a search query, filters, group by aggregations, accounts to search, and trigger conditions.

    query string

    Default value: *

    Provide a Kibana search query written in Lucene syntax. The search query together with the filters select for the relevant logs.

    Cannot be null - send an asterisk wildcard * if not using a search query.

    filters object

    Apply must and must_not filters to the monitoring alert. Filters are more efficient compared to a query, so it's recommended to opt for a filter over a query, where possible. See Elasticsearch Bool-Query for more detail.

    bool object

    Runs Elasticsearch Bool Query filters on the data (before the search query is applied). The most efficient way to grab the logs you are looking for.

    must object[]
  • Array [
    • match_phrase object
    Field object
    query string
  • ]
    • must_not object[]
  • Array [
    • match_phrase object
    Field object
    query string
  • ]
    • groupBy string[]nullable

    Possible values: <= 3

    Specify 1-3 fields by which to group the results and count them. If you apply a group by operation, the alert returns a count of the results aggregated by unique values.

    aggregation object

    Specifies a trigger condition that acts as a threshold.

    aggregationType string

    Possible values: [SUM, MIN, MAX, AVG, COUNT, UNIQUE_COUNT, NONE, PERCENTAGE, PERCENTILE]

    Specifies the aggregation operator.

    • If COUNT, fieldToAggregateOn must be null, and groupBy fields must not be empty.

    • If NONE, fieldToAggregateOn must be null, and groupBy field must not be empty (or null).

    • If PERCENTAGE, valueToAggregateOn must be specified.

    • If any other operator type (other than NONE or COUNT), fieldToAggregateOn must not be null.

    fieldToAggregateOn string

    Selects the field on which to run the aggregation for the trigger condition.

    • Cannot be a field already in use for groupBy.
    valueToAggregateOn string

    Used by the PERCENTAGE aggregation to select the field’s value. This value is used to determine if its ratio out of the total amount of logs in the query satisfies the trigger condition.

    • Only relevant for the PERCENTAGE aggregation.
    shouldQueryOnAllAccounts boolean

    Default value: true

    Only applicable when the alert is run from the main account. If true, the alert runs on the main account and all associated searchable sub accounts. If false, specify relevant account IDs for the alert to monitor using the accountIdsToQueryOn field.

    accountIdsToQueryOn int32[]

    Specify Account IDs to select which accounts the alert should monitor. The alert will be checked only on these accounts.

    trigger object

    Sets the triggering threshold and severity tab to label the event when the alert triggers.

    operator string

    Possible values: [LESS_THAN, GREATER_THAN, LESS_THAN_OR_EQUALS, GREATER_THAN_OR_EQUALS, EQUALS, NOT_EQUALS]

    Specifies the operator for evaluating the results.

    severityThresholdTiers object

    Possible values: [INFO, LOW, MEDIUM, HIGH, SEVERE]

    Default value: [object Object]

    Sets a severity label per trigger threshold as a key:value pair.

    If using more than one sub-component, only 1 severityThresholdTiers is allowed. Otherwise, 1 per enum are allowed (for a total of 5 thresholds of increasing severities).

    Increasing severity must adhere to the logic of the operator.

    output object

    Selects the data output to be sent in the notification when the alert triggers. Not applicable, when grouping by fields or aggregating results, as the output is auto-selected.

    shouldUseAllFields boolean

    Default value: true

    If true, the notification output will include entire logs with all of their fields in the sample data.

  • ]
    • correlations object

    Only applicable when multiple sub-components are in use. Selects a logic for correlating the alert’s sub-components.

    AND is currently the only supported operator. When AND is the correlationOperator, both sub-components must meet their triggering criteria for the alert to trigger.

    correlationOperators string[]

    Possible values: [AND]

    joins object[]

    Specifies which group by fields must have the same values to trigger the alert.

    Joins the group by fields from the first and second sub-components. The key represents the index of the sub component in the array (See the example - the index of the first sub-component is 0, the second is 1).

    The fields must be ordered pairs of the group by fields already in use in the queryDefinition.

    schedule object

    Defines the intervals in which an alert will be evaluated. This feature is still in production, but the payload already contains the data.

    cronExpression string

    Cron job for the intervals schedule.

    timezone string

    Time zone for the cron job. If no time zone is selected, UTC will be used by default.

Loading...