Important: This documentation is about an older version. It's relevant only to the release noted, many of the features and functions have been updated or replaced. Please view the current version.
k6/execution
k6/execution
provides the capability to get information about the current test execution state inside the test script. You can read in your script the execution state during the test execution and change your script logic based on the current state.
The k6/execution
module provides the test execution information with the following properties:
import exec from 'k6/execution';
export const options = {
scenarios: {
myscenario: {
// this will be the returned name
executor: 'shared-iterations',
maxDuration: '30m',
},
},
};
export default function () {
console.log(exec.scenario.name); // myscenario
}
ℹ️ Identifiers
All unique identifiers are sequentially generated starting from a base of zero (iterations) or one (VU IDs). In distributed/cloud test runs, the test-wide iteration numbers and VU identifiers are still going to be unique across instances, though there might be gaps in the sequences when, for example, some instances execute faster iterations than others or allocate more VUs mid-test.
instance
The instance object provides information associated with the load generator instance. You can think of it as the current running k6 process, which will likely be a single process if you are running k6 on your local machine. When running a cloud/distributed test with multiple load generator instances, the values of the following properties can differ across instances.
Property | Type | Description |
---|---|---|
iterationsInterrupted | integer | The number of prematurely interrupted iterations in the current instance. |
iterationsCompleted | integer | The number of completed iterations in the current instance. |
vusActive | integer | The number of active VUs. |
vusInitialized | integer | The number of currently initialized VUs. |
currentTestRunDuration | float | The time passed from the start of the current test run in milliseconds. |
scenario
Meta information and execution details about the current running scenario.
Property | Type | Description |
---|---|---|
name | string | The assigned name of the running scenario. |
executor | string | The name of the running Executor type. |
startTime | integer | The Unix timestamp in milliseconds when the scenario started. |
progress | float | Percentage in a 0 to 1 interval of the scenario progress. |
iterationInInstance | integer | The unique and zero-based sequential number of the current iteration in the scenario, across the current instance. |
iterationInTest | integer | The unique and zero-based sequential number of the current iteration in the scenario. It is unique in all k6 execution modes - in local, cloud and distributed/segmented test runs. However, while every instance will get non-overlapping index values in cloud/distributed tests, they might iterate over them at different speeds, so the values won’t be sequential across them. |
test
Control the test execution.
Property | Type | Description |
---|---|---|
abort([String]) | function | It aborts the test run with the exit code 108 , and an optional string parameter can provide an error message. Aborting the test will not prevent the teardown() execution. |
options | Object | It returns an object with all the test options as properties. The options’ values are consolidated following the order of precedence and derived if shortcuts have been used. It returns null for properties where the relative option hasn’t been defined. |
vu
Meta information and execution details about the current vu.
Property | Type | Description |
---|---|---|
iterationInInstance | integer | The identifier of the iteration in the current instance for this VU. This is only unique for current VU and this instance (if multiple instances). This keeps being aggregated if a given VU is reused between multiple scenarios. |
iterationInScenario | integer | The identifier of the iteration in the current scenario for this VU. This is only unique for current VU and scenario it is currently executing. |
idInInstance | integer | The identifier of the VU across the instance. Not unique across multiple instances. |
idInTest | integer | The globally unique (across the whole test run) identifier of the VU. |
metrics.tags | object | The map that gives control over VU’s Tags. The Tags will be included in every metric emitted by the VU and the Tags’ state is maintained across different iterations of the same Scenario while the VU exists. Check how to use it in the example below. |
metrics.metadata | object | The map that gives control over VU’s Metadata. The Metadata will be included in every metric emitted by the VU and the Metadata’s state is maintained across different iterations of the same Scenario while the VU exists. Check how to use it in the example below. |
Setting a Tag with the same key as a system tag is allowed, but it requires attention to avoid unexpected results. Overwriting system tags will not throw an error, but in most cases will not actually change the value of the emitted metrics as expected. For example, trying to set the url
tag value will not result in a changed tag value when http.get()
is called, since the tag value is determined by the HTTP request itself. However, it will add the tag url
to the metric samples emitted by a check()
or metric.add()
, which is probably not the desired behavior. On the other hand, setting the name
tag will work as expected, since that was already supported for http.*
methods, for the purposes of the URL Grouping feature.
Not all the types are accepted as a tag value: k6 supports strings, numbers and boolean types. Under the hood, the tags
object handles a Tag as a String
key-value pair, so all the types will be implicitly converted into a string. If one of the denied types is used (e.g. Object or Array) and the throw
option is set, an exception will be thrown. Otherwise, only a warning is printed and the tag value will be discarded.
Examples and use cases
Getting unique data once
This is a common use case for data parameterization, you can read the examples using scenario.iterationInTest
and vu.idInTest
.
Timing operations
The startTime
property from the scenario
object can be used to time operations.
import exec from 'k6/execution';
export default function () {
// do some long operations
// ...
console.log(`step1: scenario ran for ${new Date() - new Date(exec.scenario.startTime)}ms`);
// some more long operations
//...
console.log(`step2: scenario ran for ${new Date() - new Date(exec.scenario.startTime)}ms`);
}
Script naming
The name
property can be used for executing the logic based on which script is currently running.
Tip: If you need to run multiple scenarios in your script you can use
exec
option achieve a similar goal
import exec from 'k6/execution';
export const options = {
scenarios: {
'the-first': {
// ...
},
'the-second': {
// ...
},
},
};
export default function () {
if (exec.scenario.name === 'the-first') {
// do some logic during this scenario
} else {
// do some other logic in the others
}
}
Test Abort
Aborting is possible during initialization:
import exec from 'k6/execution';
exec.test.abort();
As well as inside the default
function:
import exec from 'k6/execution';
export default function () {
// Note that you can abort with a specific message too
exec.test.abort('this is the reason');
}
export function teardown() {
console.log('teardown will still be called after test.abort()');
}
Get test options
Get the consolidated and derived options’ values
import exec from 'k6/execution';
export const options = {
stages: [
{ duration: '5s', target: 100 },
{ duration: '5s', target: 50 },
],
};
export default function () {
console.log(exec.test.options.paused); // null
console.log(exec.test.options.scenarios.default.stages[0].target); // 100
}
Tags
The vu.metrics.tags
property can be used for getting or setting VU’s tags.
import http from 'k6/http';
import exec from 'k6/execution';
export default function () {
exec.vu.metrics.tags['mytag'] = 'value1';
exec.vu.metrics.tags['mytag2'] = 2;
// the metrics these HTTP requests emit will get tagged with `mytag` and `mytag2`:
http.batch(['https://test.k6.io', 'https://test-api.k6.io']);
}
vu.tags
(without metrics
) can also be used, but is deprecated for the more context-specific variant.
Metadata
The vu.metrics.metadata
property can be used for getting or setting VU’s metadata. It is similar to tags
, but can be used for high cardinality data. It also can not be used in thresholds and will likely be handled differently by each output.
import http from 'k6/http';
import exec from 'k6/execution';
export default function () {
exec.vu.metrics.metadata['trace_id'] = 'somecoolide';
// the metrics these HTTP requests emit will get the metadata `trace_id`:
http.batch(['https://test.k6.io', 'https://test-api.k6.io']);
delete exec.vu.metrics.metadata['trace_id']; // this will unset it
// which will make the metrics these requests to not have the metadata `trace_id` set on them.
http.batch(['https://test.k6.io', 'https://test-api.k6.io']);
}