Template Variable Evaluation
In monitor notification messages, you can modify the output of template variables using the eval syntax, which enables several different mathematical operations and functions on template variables with a numerical or string value.
Operators
Evaluation operators allow you to perform basic arithmetic operations on a numerical template variable. The syntax uses the following format. Note: The expressions must be wrapped in quotation marks (").
{{eval "TEMPLATE_VARIABLE_NAME+1-2*3/4"}}
The following operators are supported:
| Operator | Description |
|---|
| + | Addition |
| - | Subtraction |
| * | Multiplication |
| / | Division |
| ^ | Exponentiation |
| % | Modulo |
Example
The {{last_triggered_at_epoch}} template variable returns the UTC time when a monitor last triggered in milliseconds epoch format.
Scope links to specific times
Evaluation operators can be used to subtract 15 minutes (15 * 60 * 1000 milliseconds) with the following:
{{eval "last_triggered_at_epoch-15*60*1000"}}
This is useful for creating time-scoped links in your monitor notification message to other pages in Datadog or other workflow tools. For instance, use the evaluation operators on {{last_triggered_at_epoch}} to create a time-scoped link to the Datadog Log Explorer:
https://app.datadoghq.com/logs?from_ts={{eval "last_triggered_at_epoch-15*60*1000"}}&to_ts={{last_triggered_at_epoch}}&live=false
Routing notifications to different teams based on time of day
You can combine a modulo % evaluation of the last_triggered_at_epoch variable with {{#is_exact_match}}{{/is_exact_match}} to customize the routing of notifications based on time of day (UTC):
{{#is_exact_match (eval "int(last_triggered_at_epoch / 3600000 % 24)") "8" "9" "10" "11" "12" "13"}}
Handle that should receive notification if time is between 8AM and 2PM UTC
{{/is_exact_match}}
Note: If you need to evaluate your monitor on a schedule, see Custom Schedules instead.
Functions
The value of a template variable can be used as the input for evaluation functions to change the formatting of the template variable or perform a mathematical operation on the value (if applicable). The syntax uses the following format. Note: The expressions must be wrapped in quotation marks (").
{{eval "function(TEMPLATE_VARIABLE_NAME)"}}
The following functions change how the value of a numerical template variable is formatted:
| Function | Description |
|---|
| humanize_bytes(var) | Returns a human readable formatting of var bytes |
| humanize_bits(var) | Returns a human readable formatting of var bits |
| abs(var) | Returns the absolute value of var |
| int(var) | Returns var as a floored integer; that is, the numbers to the left of the decimal point. For example: if var = 12.345, then int(var) returns 12. |
| float(var) | Returns var as a float |
| trunc(var) | Returns var as an integer, an alias of the int function |
| dec(var) | Returns the numbers to the right of the decimal point. For example: if var = 12.345, then dec(var) returns 0.345. |
The following functions use the value of a numerical template variable as the input to a mathematical function:
| Function | Description |
|---|
| round(var) | Returns var rounded to the nearest integer |
| round(var, n) | Returns var rounded to a specified number of digits (n).
For example: round(12.376, 2) = 12.38 |
| ceil(var) | Returns the ceiling of var (the smallest integer that is greater than or equal to var) |
| floor(var) | Returns the floor of var (the largest integer that is less than or equal to var) |
| sgn(var) | Returns the value of the sign function evaluated at var:
sgn(var) = 1 if var > 0
sgn(var) = 0 if var = 0
sgn(var) = -1 if var < 0 |
| to_bool(var) | Returns true if var = 1
Returns false if var = 0 |
| exp(var) | Returns e (the base of the natural logarithm) raised to the power of var |
| log10(var) | Returns the base-10 logarithm of var |
| sin(var) | Returns the sine of var radians |
| sinh(var) | Returns the hyperbolic sine of var |
| asin(var) | Returns the arc sine of var, in radians |
| asinh(var) | Returns the inverse hyperbolic sine of var |
| cos(var) | Returns the cosine of var radians |
| cosh(var) | Returns the hyperbolic cosine of var |
| acos(var) | Returns the arc cosine of var, in radians |
| acosh(var) | Returns the inverse hyperbolic cosine of var |
| tan(var) | Returns the tangent of var radians |
| tanh(var) | Returns the hyperbolic tangent of var |
| atan(var) | Returns the arc tangent of var, in radians |
| atan2(var1, var2) | Returns atan(var1 / var2), in radians |
| atanh(var) | Returns the inverse hyperbolic tangent of var |
Examples
If the decimal places of the {{value}} template variable are unnecessary for your particular use case, use the int function to evaluate {{value}} as an integer to improve readability and remove the decimals:
If {{value}} is evaluating to a large number of bytes or bits, use the humanize_bytes or humanize_bits function to convert the number to a different higher order memory unit like GB or MB to improve readability:
{{eval "humanize_bytes(value)"}}
{{eval "humanize_bits(value)"}}
The following functions can be used to perform certain operations on strings variables:
| Function | Description |
|---|
| upper(var) | Returns a string converted to uppercase letters |
| lower(var) | Returns a string converted to lowercase letters |
| substring(var, start, end) | Extracts characters from a string, between two specified indexes (start, end). The third parameter is optional.
For example: substring(“host:D”, 5) = “D” |
| strip(var, characters) | Remove the leading and trailing characters. When the second parameter is null, it removes spaces at the beginning and at the end of the string. For example:
strip(" host:E “) = “host:E”
strip(“abchost:Eabc”, “abc”) = “host:E” |
