Library with a bunch of useful functions that extend Gatling DSL and make your performance better.
If you are using TinkoffCreditSystems/gatling-template.g8, you already have all dependencies in it. Gatling Template Project
Add dependency with version that you need
libraryDependencies += "ru.tinkoff" %% "gatling-picatinny" % "<version>"
The only class that you need from this module is SimulationConfig
. It could be used to attach some default variables
such as intensity
, baseUrl
, baseAuthUrl
and some others to your scripts. Also, it provides functions to get custom
variables fom config.
Import:
Scala example:
import ru.tinkoff.gatling.config.SimulationConfig._
Java example:
import static ru.tinkoff.gatling.javaapi.SimulationConfig.*;
Kotlin example:
import ru.tinkoff.gatling.javaapi.SimulationConfig.*
Using default variables:
Scala example:
import ru.tinkoff.gatling.config.SimulationConfig._
val testPlan: Seq[OpenInjectionStep] = List(
rampUsersPerSec(0).to(intensity).during(rampDuration),
constantUsersPerSec(intensity).during(stageDuration)
)
Java example:
import static ru.tinkoff.gatling.javaapi.SimulationConfig.*;
SomeScenario.scn.injectOpen(
incrementUsersPerSec(intensity() / stagesNumber())
.times(stagesNumber())
.eachLevelLasting(stageDuration())
.separatedByRampsLasting(rampDuration())
.startingFrom(0)
)
Kotlin example:
import ru.tinkoff.gatling.javaapi.SimulationConfig.*
SomeScenario.scn.injectOpen(
incrementUsersPerSec(intensity() / stagesNumber())
.times(stagesNumber())
.eachLevelLasting(stageDuration())
.separatedByRampsLasting(rampDuration())
.startingFrom(0.0),
)
Using functions to get custom variable:
simulation.conf
stringVariable: "FOO",
intVariable: 1,
doubleVariable: 3.1415,
duration: {
durationVariable: 3600s
}
booleanVariable: true
Scala example:
import ru.tinkoff.gatling.config.SimulationConfig._
val stringVariable = getStringParam("stringVariable")
val intVariable = getIntParam("intVariable")
val doubleVariable = getDoubleParam("doubleVariable")
val durationVariable = getDurationParam("duration.durationVariable")
val booleanVariable = getBooleanParam("booleanVariable")
Java example:
import static ru.tinkoff.gatling.javaapi.SimulationConfig.*;
String stringVariable = getStringParam("stringVariable");
int intVariable = getIntParam("intVariable");
double doubleVariable = getDoubleParam("doubleVariable");
Duration durationVariable = getDurationParam("duration.durationVariable");
boolean booleanVariable = getBooleanParam("booleanVariable");
kotlin example:
import ru.tinkoff.gatling.javaapi.SimulationConfig.*
val stringVariable = getStringParam("stringVariable")
val intVariable = getIntParam("intVariable")
val doubleVariable = getDoubleParam("doubleVariable")
val durationVariable = getDurationParam("duration.durationVariable")
val booleanVariable = getBooleanParam("booleanVariable")
This module contains vast number of random feeders. They could be used as regular feeders and realize common needs, i.e. random phone number or random digit. Now it supports feeders for dates, numbers and digits, strings, uuids, phones. Basic examples will be provided below. Other feeders can be used in a similar way.
Scala example:
import ru.tinkoff.gatling.feeders._
//creates feeder with name 'randomString' that gets random string of length 10
val stringFeeder = RandomStringFeeder("randomString", 10)
//creates feeder with name 'digit' that gets random Int digit
val digitFeeder = RandomDigitFeeder("digit")
//creates feeder with name 'uuid' that gets random uuid
val uuidFeeder = RandomUUIDFeeder("uuid")
Java example:
import static ru.tinkoff.gatling.javaapi.Feeders.*;
//creates feeder with name 'randomString' that gets random string of length 10
Iterator<Map<String, Object>> stringFeeder = RandomStringFeeder("randomString", 10);
//creates feeder with name 'digit' that gets random Int digit
Iterator<Map<String, Object>> digitFeeder = RandomDigitFeeder("digit");
//creates feeder with name 'uuid' that gets random uuid
Iterator<Map<String, Object>> uuidFeeder = RandomUUIDFeeder("uuid");
Kotlin example:
import ru.tinkoff.gatling.javaapi.Feeders.*
//creates feeder with name 'randomString' that gets random string of length 10
val stringFeeder = RandomStringFeeder("string", 10)
//creates feeder with name 'digit' that gets random Int digit
val digitFeeder = RandomDigitFeeder("digit")
//creates feeder with name 'uuid' that gets random uuid
val uuidFeeder = RandomUUIDFeeder("uuid")
Creates feeder capable of retrieving secret data from HC Vault
- authorises via approle;
- uses v1 API;
- works with kv Secret Engine;
- does not iterate over keys, returns full map with keys it found on each call;
- params:
- vaultUrl - vault URL e.g. "https://vault.ru"
- secretPath - path to secret data within your vault e.g. "testing/data"
- roleId - approle login
- secretId - approle password
- keys - list of keys you are willing to retrieve from vault
Scala example:
val vaultFeeder = VaultFeeder(vaultUrl, secretPath, roleId, secretId, keys)
Java example:
Iterator<Map<String, Object>> vaultFeeder = VaultFeeder(vaultUrl, secretPath, roleId, secretId, keys);
Kotlin example:
val vaultFeeder = VaultFeeder(vaultUrl, secretPath, roleId, secretId, keys)
Creates a feeder with separated values from a source String, Seq[String] or Seq[Map[String, Any]].
- params:
- paramName - feeder name
- source - data source
- separator - ",", ";", "\t" or other delimiter which separates values. You can also use following methods for the most common separators: .csv(...), .ssv(...), .tsv(...)
Get separated values from a source: String
Scala example:
val sourceString = "v21;v22;v23"
val separatedValuesFeeder: FeederBuilderBase[String] =
SeparatedValuesFeeder("someValues", sourceString, ';') // Vector(Map(someValues -> v21), Map(someValues -> v22), Map(someValues -> v23))
Java example:
String sourceString = "v21;v22;v23";
Iterator<Map<String, Object>> separatedValuesFeeder = SeparatedValuesFeeder.apply("someValues", sourceString, ';');
Kotlin example:
val sourceString = "v21;v22;v23"
val separatedValuesFeeder = SeparatedValuesFeeder.apply("someValues", sourceString, ';')
Get separated values from a source: Seq[String]
Scala example:
val sourceSeq = Seq("1,two", "3,4")
val separatedValuesFeeder: FeederBuilderBase[String] =
SeparatedValuesFeeder.csv("someValues", sourceSeq) // Vector(Map(someValues -> 1), Map(someValues -> two), Map(someValues -> 3), Map(someValues -> 4))
Java example:
List<Map<String, Object>> sourceList = Arrays.asList("1,two", "3,4");
Iterator<Map<String, Object>> separatedValuesFeeder = SeparatedValuesFeeder.csv("someValues", sourceList);
Kotlin example:
var sourceList = listOf("1,two", "3,4")
var separatedValuesFeeder1 = SeparatedValuesFeeder.csv("someValues", sourceList)
Get separated values from a source: Seq[Map[String, Any]]
Scala example:
val vaultFeeder: FeederBuilderBase[String] = Vector(
Map(
"HOSTS" -> "host11,host12",
"USERS" -> "user11",
),
Map(
"HOSTS" -> "host21,host22",
"USERS" -> "user21,user22,user23",
),
)
val separatedValuesFeeder: FeederBuilderBase[String] =
SeparatedValuesFeeder(None, vaultFeeder.readRecords, ',') // Vector(Map(HOSTS -> host11), Map(HOSTS -> host12), Map(USERS -> user11), Map(HOSTS -> host21), Map(HOSTS -> host22), Map(USERS -> user21), Map(USERS -> user22), Map(USERS -> user23))
Java example:
List<Map<String, Object>> vaultData = Arrays.asList(Map.of("HOSTS","host11,host12"), Map.of("USERS", "user21,user22,user23"));
Iterator<Map<String, Object>> separatedValuesFeeder = SeparatedValuesFeeder.apply(Optional.empty(), vaultData, ',');
Kotlin example:
var sourceList = listOf(Map.of("HOSTS", "host11,host12"), Map.of("USERS", "user21,user22,user23"))
var separatedValuesFeeder1 = SeparatedValuesFeeder.csv(null, sourceList)
Creates a feeder with phone numbers with formats from json file or case class PhoneFormat
Simple phone feeder
Scala example:
val simplePhoneNumber: Feeder[String] = RandomPhoneFeeder("simplePhoneFeeder")
Java example:
Iterator<Map<String, Object>> simplePhoneNumber = RandomPhoneFeeder("simplePhoneFeeder");
Kotlin example:
val simplePhoneNumber = RandomPhoneFeeder("simplePhoneNumber")
Phone feeder with custom formats
Scala example:
val ruMobileFormat: PhoneFormat = PhoneFormat(
countryCode = "+7",
length = 10,
areaCodes = Seq("903", "906", "908"),
prefixes = Seq("55", "81", "111"),
format = "+X XXX XXX-XX-XX")
val randomPhoneNumber: Feeder[String] =
RandomPhoneFeeder("randomPhoneNumber", ruMobileFormat)
Java example:
PhoneFormat ruMobileFormat = PhoneFormatBuilder.apply("+7", 10, Arrays.asList("945", "946"), "+X XXX XXX-XX-XX", Arrays.asList("55", "81", "111"));
Iterator<Map<String, Object>> randomPhoneNumber = RandomPhoneFeeder("randomPhoneNumber", ruMobileFormat);
Kotlin example:
val ruMobileFormat = PhoneFormatBuilder.apply(
"+7",
10,
listOf("945", "946"),
"+X XXX XXX-XX-XX",
listOf("55", "81", "111")
)
val randomPhoneNumber = RandomPhoneFeeder("randomPhoneNumber", ruMobileFormat)
Phone feeder with custom formats with file
Creates file with formats, for example RESOURCES/phoneTemplates/ru.json
{
"formats": [
{
"countryCode": "+7",
"length": 10,
"areaCodes": ["903", "906"],
"prefixes": ["123", "321", "132", "231"],
"format": "+X(XXX)XXXXXXX"
},
{
"countryCode": "8",
"length": 10,
"areaCodes": ["495", "499"],
"prefixes": ["81", "82", "83"],
"format": "X(XXX)XXX-XX-XX"
}
]
}
Scala example:
val phoneFormatsFromFile: String = "phoneTemplates/ru.json"
val randomE164PhoneNumberFromJson: Feeder[String] =
RandomPhoneFeeder("randomE164PhoneNumberFile", phoneFormatsFromFile, TypePhone.E164PhoneNumber)
Java example:
String phoneFormatsFromFile = "phoneTemplates/ru.json";
Iterator<Map<String, Object>> randomE164PhoneNumberFromJson = RandomPhoneFeeder("randomE164PhoneNumberFile", phoneFormatsFromFile, TypePhone.E164PhoneNumber());
Kotlin example:
val phoneFormatsFromFile = "phoneTemplates/ru.json"
val randomE164PhoneNumberFromJson = RandomPhoneFeeder("randomE164PhoneNumberFile", phoneFormatsFromFile, TypePhone.E164PhoneNumber())
This module allows you to write custom points to InfluxDB.
- Write Start/Stop annotations before/after simulation run
- Write custom points to InfluxDB
Scala example:
Import:
import ru.tinkoff.gatling.influxdb.Annotations
For using you should simply add with Annotations
for your simulation class:
class LoadTest extends Simulation with Annotations {
//some code
}
Java example:
Import:
import ru.tinkoff.gatling.javaapi.influxdb.SimulationWithAnnotations;
For using you should extend your simulation class from SimulationWithAnnotations
:
class LoadTest extends SimulationWithAnnotations {
//some code
}
Kotlin example:
Import:
import ru.tinkoff.gatling.javaapi.influxdb.SimulationWithAnnotations
For using you should extend your simulation class from SimulationWithAnnotations
:
class LoadTest : SimulationWithAnnotations(){
//some code
}
To see annotations in Grafana you need this two queries, where Perfix
is from gatling.data.graphite.rootPathPrefix
in gatling.conf
:
SELECT "annotation_value" FROM "${Prefix}" where "annotation" = 'Start'
SELECT "annotation_value" FROM "${Prefix}" where "annotation" = 'Stop'
!DANGER! Read before use:
- Not intended for load testing of InfluxDB.
- You can easily waste InfluxDB with junk data. Don't use frequently changing keys.
- When recording points in the setUp simulation, a separate script will be created, which will be displayed in the test status in the console and in the final Gatling data.
- Depending on your settings, Gatling will write simulation data to InfluxDB in batches every n seconds. In this case, the timestamp of the custom point will be taken during its recording, which may cause inaccuracies when displaying data.
Scala example:
Import:
import ru.tinkoff.gatling.influxdb.Annotations._
Using:
//if default prepared Point doesn't suit you
Point(configuration.data.graphite.rootPathPrefix, System.currentTimeMillis() * 1000000)
.addTag(tagName, tagValue)
.addField(fieldName, fieldValue)
//prepare custom Point*
import io.razem.influxdbclient.Point
def customPoint(tag: String, value: String) = Point(configuration.data.graphite.rootPathPrefix, System.currentTimeMillis() * 1000000)
.addTag("myTag", tag)
.addField("myField", value) //value: Boolean | String | BigDecimal | Long | Double
//write custom prepared Point from scenario
.exec(
...)
.userDataPoint(customPoint("custom_tag", "inside_scenario"))
.exec(
...)
//write default prepared Point from scenario
.exec(
...)
.userDataPoint("myTag", "tagValue", "myField", "fieldValue")
//also you can use gatling Expression language for values (could waste DB):
.userDataPoint("myTag", "${variableFromGatlingSession}", "myField", "${anotherVariableFromSession}")
.exec(
...)
//write Point from Simulation setUp
setUp(
firstScenario.inject(atOnceUsers(1))
//write point after firstScenario execution
//"write_first_point" the name of the scenario, will be displayed in the simulation stats
.userDataPoint("write_first_point", customPoint("custom_tag", "between_scenarios"))
.andThen(
secondScenario.inject(atOnceUsers(1))
//similar to simple .userDataPoint, write point after secondScenario execution
.andThen(
userDataPoint("write_second_point", "custom_tag", "after_second", "custom_field", "end")
)
)
)
Java example:
Import:
import static ru.tinkoff.gatling.javaapi.influxdb.Annotations.*;
Using:
//if default prepared Point doesn't suit you
Point(configuration.data.graphite.rootPathPrefix, System.currentTimeMillis() * 1000000)
.addTag(tagName, tagValue)
.addField(fieldName, fieldValue)
//prepare custom Point*
import io.razem.influxdbclient.Point;
static Point customPoint(tag: String, value: String){ return Point(configuration.data.graphite.rootPathPrefix,System.currentTimeMillis()*1000000)
.addTag("myTag",tag)
.addField("myField",value) //value: Boolean | String | BigDecimal | Long | Double
}
//write custom prepared Point from scenario
.exec(
...)
.exec(userDataPoint(customPoint("custom_tag", "inside_scenario")))
.exec(
...)
//write default prepared Point from scenario
.exec(
...)
.exec(userDataPoint("myTag", "tagValue", "myField", "fieldValue"))
//also you can use gatling Expression language for values (could waste DB):
.exec(userDataPoint("myTag", "${variableFromGatlingSession}", "myField", "${anotherVariableFromSession}"))
.exec(
...)
//write Point from Simulation setUp
setUp(
firstScenario.inject(atOnceUsers(1))
//write point after firstScenario execution
//"write_first_point" the name of the scenario, will be displayed in the simulation stats
.andThen(userDataPoint("write_first_point", customPoint("custom_tag", "between_scenarios")))
.andThen(
secondScenario.inject(atOnceUsers(1))
//similar to simple .userDataPoint, write point after secondScenario execution
.andThen(
userDataPoint("write_second_point", "custom_tag", "after_second", "custom_field", "end")
)
)
)
Kotlin example:
Import:
import ru.tinkoff.gatling.javaapi.influxdb.Annotations.*
Using:
//if default prepared Point doesn't suit you
Point(configuration.data.graphite.rootPathPrefix, System.currentTimeMillis() * 1000000)
.addTag(tagName, tagValue)
.addField(fieldName, fieldValue)
//prepare custom Point*
import io.razem.influxdbclient.Point
fun customPoint(tag: String, value: String){ Point(configuration.data.graphite.rootPathPrefix,System.currentTimeMillis()*1000000)
.addTag("myTag",tag)
.addField("myField",value) //value: Boolean | String | BigDecimal | Long | Double
}
//write custom prepared Point from scenario
.exec(
...)
.exec(userDataPoint(customPoint("custom_tag", "inside_scenario")))
.exec(
...)
//write default prepared Point from scenario
.exec(
...)
.exec(userDataPoint("myTag", "tagValue", "myField", "fieldValue"))
//also you can use gatling Expression language for values (could waste DB):
.exec(userDataPoint("myTag", "${variableFromGatlingSession}", "myField", "${anotherVariableFromSession}"))
.exec(
...)
//write Point from Simulation setUp
setUp(
firstScenario.inject(atOnceUsers(1))
//write point after firstScenario execution
//"write_first_point" the name of the scenario, will be displayed in the simulation stats
.andThen(userDataPoint("write_first_point", customPoint("custom_tag", "between_scenarios")))
.andThen(
secondScenario.inject(atOnceUsers(1))
//similar to simple .userDataPoint, write point after secondScenario execution
.andThen(
userDataPoint("write_second_point", "custom_tag", "after_second", "custom_field", "end")
)
)
)
- Load profile configs from HOCON or YAML files
- Common traits to create profiles for any protocol
- HTTP profile as an example
import ru.tinkoff.gatling.profile._
import pureconfig.generic.auto._
HOCON configuration example:
{
name: test-profile
profile: [
{
name: request-a
url: "http://test.url"
probability: 50.0
method: get
},
{
name: request-b
url: "http://test.url"
probability: 50.0
method: post
body: "{\"a\": \"1\"}"
}
]
}
YAML configuration example:
name: test-profile
profile:
- name: request-a
url: "http://test.url"
probability: 50
method: get
- name: request-b
url: "http://test.url"
probability: 50
method: post
body: "{\"a\": \"1\"}"
Simulation setUp
class test extends Simulation {
val profileConfigName = "profile.conf"
val someTestPlan = constantUsersPerSec(intensity) during stageDuration
val httpProtocol = http.baseUrl(baseUrl)
val config: HttpProfileConfig = new ProfileBuilder[HttpProfileConfig].buildFromYaml(profileConfigName)
val scn: ScenarioBuilder = config.toRandomScenario
setUp(
scn.inject(
atOnceUsers(10)
).protocols(httpProtocol)
).maxDuration(10)
}
New profile YAML configuration example:
apiVersion: link.ru/v1alpha1
kind: PerformanceTestProfiles
metadata:
name: performance-test-profile
description: performance test profile
spec:
profiles:
- name: maxPerf
period: 10.05.2022 - 20.05.2022
protocol: http
profile:
- request: request-1
intensity: 100 rph
groups: ["Group1"]
params:
method: POST
path: /test/a
headers:
- 'Content-Type: application/json'
- 'Connection: keep-alive'
body: '{"a": "b"}'
- request: request-2
intensity: 200 rph
groups: ["Group1", "Group2"]
params:
method: GET
path: /test/b
body: '{"c": "d"}'
- request: request-3
intensity: 200 rph
groups: [ "Group1", "Group2" ]
params:
method: GET
path: /test/c
body: '{"e": "f"}'
Optional fields: groups, headers, body.
If there are no required fields, an exception will be thrown for the missing field.
Simulation setUp
Scala example:
class Debug extends Simulation {
val pathToProfile = "path/to/profile.yml"
val scn = ProfileBuilderNew.buildFromYaml(pathToProfile).selectProfile("maxPerf").toRandomScenario
setUp(
scn.inject(
atOnceUsers(10)
).protocols(httpProtocol)
)
.maxDuration(10)
}
Java example:
import ru.tinkoff.gatling.javaapi.profile.ProfileBuilderNew;
public class Debug extends Simulation {
public static ScenarioBuilder scn = ProfileBuilderNew
.buildFromYaml("path/to/profile.yml")
.selectProfile("maxPerf")
.toRandomScenario();
{
setUp(
scn.injectOpen(atOnceUsers(1))
).protocols(httpProtocol);
}
}
Kotlin example:
import ru.tinkoff.gatling.javaapi.profile.ProfileBuilderNew
class Debug : Simulation() {
val scn: ScenarioBuilder = ProfileBuilderNew
.buildFromYaml("path/to/profile.yml")
.selectProfile("maxPerf")
.toRandomScenario()
init {
setUp(
scn.injectOpen(atOnceUsers(1)),
).protocols(httpProtocol)
}
}
This module allows you to use Redis commands.
- Support Redis commands: SADD, DEL, SREM
- Support Gatling EL
- Мethods are not taken into account in statistics Gatling.
- Not intended for load testing of Redis.
Scala example:
import com.redis.RedisClientPool
import ru.tinkoff.gatling.redis.RedisActionBuilder._
Java example:
import io.gatling.javaapi.redis.RedisClientPool;
import ru.tinkoff.gatling.javaapi.redis.RedisClientPoolJava;
Kotlin example:
import io.gatling.javaapi.redis.RedisClientPool
import ru.tinkoff.gatling.javaapi.redis.RedisClientPoolJava
First you need to prepare RedisClientPool:
Scala example:
val redisPool = new RedisClientPool(redisUrl, 6379)
Java example:
static RedisClientPool redisClientPool = new RedisClientPool("localhost", 6379);
static RedisClientPoolJava redisClientPoolJava = new RedisClientPoolJava(redisClientPool);
//or
static RedisClientPoolJava redisClientPoolJava = new RedisClientPoolJava("localhost", 6379);
Kotlin example:
val redisClientPool = RedisClientPool("localhost", 6379)
val redisClientPoolJava = RedisClientPoolJava(redisClientPool)
//or
val redisClientPoolJava = RedisClientPoolJava("localhost", 6379)
Add the Redis commands to your scenario chain:
Scala example:
.exec(redisPool.SADD("key", "values", "values")) //add the specified members to the set stored at key
.exec(redisPool.DEL("key", "keys")) //removes the specified keys
.exec(redisPool.SREM("key", "values", "values")) //remove the specified members from the set stored at key
Java example:
.exec(redisClientPoolJava.SADD("key", "values", "values")) //add the specified members to the set stored at key
.exec(redisClientPoolJava.DEL("key", "keys")) //removes the specified keys
.exec(redisClientPoolJava.SREM("key", "values", "values")) //remove the specified members from the set stored at key
Kotlin example:
.exec(redisClientPoolJava.SADD("key", "values", "values")) //add the specified members to the set stored at key
.exec(redisClientPoolJava.DEL("key", "keys")) //removes the specified keys
.exec(redisClientPoolJava.SREM("key", "values", "values")) //remove the specified members from the set stored at key
This module contains some syntax extensions for http requests with json body. It allows embed json-body in request
with jsonBody
method for HttpRequestBuilder
. And this module is provided ability to send request body templates from
files in resource subfolder resources/templates
by filename. Sending of templates may be done with
method postTemplate
from trait Templates
This part contains http request Json body DSL.
For use this you need import this:
import ru.tinkoff.gatling.templates.HttpBodyExt._
import ru.tinkoff.gatling.templates.Syntax._
Then use described later constructions for embed jsonBody in http requests. For example, you write something like this:
class SampleScenario {
val sendJson: ScenarioBuilder =
scenario("Post some")
.exec(
http("PostData")
.post(url)
.jsonBody(
"id" - 23, // in json - "id" : 23
"name", // in json it interpreted as - "name" : get value from session variable ${name}
"project" - ( // in json - "project" : { ... }
"id" ~ "projectId", // in json - "id" : value from session var ${projectId}
"name" - "Super Project", // in json - "name": "Super Project"
"sub" > (1, 2, 3, 4, 5, 6) // in json - "sub" : [ 1,2,3,4,5,6 ]
)
)
)
}
As result this scenario sent POST request with body:
{
"id": 23,
"name": "Test",
"project": {
"id": 23421,
"name": "Super Project",
"sub": [
1,
2,
3,
4,
5,
6
]
}
}
As you can see in the example:
- construction
"some_name" - <val>
map to"some_name": <val>
in json; - construction
"varName"
map to"varName" : <get value from session variable ${varName}>
in json; - construction
"some_name" ~ "sesVar"
map to"some_name" : <value from session var ${sesVar}>
in json; "some_name" > (<...items>)
map to array field"some_name": [ ...items ]
in json;"some_name" - (<...fields>)
map to object field"some_name": { ...fields }
in json;
Suppose in folder resources/templates contains this:
$ tree resources/
.
├── gatling.conf
├── logback.xml
├── pools
│ └── example_pool.csv
├── simulation.conf
└── templates
└── example_template1.json
└── example_template2.json
For use templates in resources/templates
you need import trait Templates
.
import ru.tinkoff.gatling.templates.Templates._
Then add this trait to your Scenario and use postTemplate
method like show later:
class SampleScenario extends Templates {
val sendTemplates: ScenarioBuilder =
scenario("Templates scenario")
.exec(postTemplate("example_template1", "/post_route"))
.exec(postTemplate("example_template2", "/post_route"))
}
This Scenario will send 2 post requests one with body from example_template1.json
, second with body
from example_template2.json
to route $baseUrl/post_route
. In template files you may use
gatling expression syntax.
- Generates a JWT token using a json template and stores it in a Gatling session, then you can use it to sign requests.
Scala:
import ru.tinkoff.gatling.utils.jwt._
Java:
import ru.tinkoff.gatling.utils.jwt.*;
import static ru.tinkoff.gatling.javaapi.utils.Jwt.*;
Kotlin
import ru.tinkoff.gatling.javaapi.utils.Jwt.*
First you need to prepare jwt generator. For example
Scala
val jwtGenerator = jwt("HS256", jwtSecretToken)
.defaultHeader
.payloadFromResource("jwtTemplates/payload.json")
Java:
static JwtGeneratorBuilder jwtGenerator = jwt("HS256", "jwtSecretToken")
.defaultHeader
.payloadFromResource("jwtTemplates/payload.json");
Kotlin
val jwtGenerator = jwt("HS256", jwtSecretToken)
.defaultHeader
.payloadFromResource("jwtTemplates/payload.json")
This will generate tokens with headers:
{
"alg": "HS256",
"typ": "JWT"
}
Payload will be generated from json template, templating is done using Gatling EL
{
"userName": "${randomString}",
"date": "${simpleDate}",
"phone": "${randomPhone}"
}
Also, the JWT generator has a DSL allowing you to (for java and kotlin similarly):
jwt("HS256", secret)
.header("""{"alg": "HS256","typ": "JWT", "customField": "customData"}""") //use custom headers from string, it must be valid json
.headerFromResource("jwtTemplates/header.json") //use src/test/resources/jwtTemplates/header.json as header template
.defaultHeader //use default jwt header, algorithm from jwt("HS256", secret), template: {"alg": "$algorithm","typ": "JWT"}
.payload("""{"userName": "${randomString}","date": "${simpleDate}","phone": "${randomPhone}"}""") //use custom payload from string, it must be valid json
.payloadFromResource("jwtTemplates/payload.json") //use src/test/resources/jwtTemplates/payload.json as payload template
For sign requests add this to your scenario chain:
.exec(_.setJwt(jwtGenerator, "jwtToken")) //generates token and save it to gatling session as "jwtToken"
.exec(addCookie(Cookie("JWT_TOKEN", "${jwtToken}").withDomain(jwtCookieDomain).withPath("/"))) //set JWT_TOKEN cookie for subsequent requests
Java:
.exec(setJwt(jwtGenerator, "jwtToken")) //generates token and save it to gatling session as "jwtToken"
.exec(addCookie(Cookie("JWT_TOKEN", "#{jwtToken}").withDomain("jwtCookieDomain").withPath("/"))) //set JWT_TOKEN cookie for subsequent requests
Kotlin
.exec(setJwt(jwtGenerator, "jwtToken")) //generates token and save it to gatling session as "jwtToken"
.exec(addCookie(Cookie("JWT_TOKEN", "#{jwtToken}").withDomain("jwtCookieDomain").withPath("/"))) //set JWT_TOKEN cookie for subsequent requests
Module helps to load assertion configs from YAML files
import ru.tinkoff.gatling.assertions.AssertionsBuilder.assertionFromYaml
File nfr contains non-functional requirements.
Requirements supported by Picatinny:
requirement | key |
---|---|
99th percentile of the responseTime | 99 перцентиль времени выполнения |
95th percentile of the responseTime | 95 перцентиль времени выполнения |
75th percentile of the responseTime | 75 перцентиль времени выполнения |
50th percentile of the responseTime | 50 перцентиль времени выполнения |
percent of the failedRequests | Процент ошибок |
maximum of the responseTime | Максимальное время выполнения |
YAML configuration example:
nfr:
- key: '99 перцентиль времени выполнения'
value:
GET /: '500'
MyGroup / MyRequest: '900'
request_1: '700'
all: '1000'
- key: 'Процент ошибок'
value:
all: '5'
- key: 'Максимальное время выполнения'
value:
GET /: '1000'
all: '2000'
Scala example
class test extends Simulation {
setUp(
scn.inject(
atOnceUsers(10)
).protocols(httpProtocol)
).maxDuration(10)
.assertions(assertionFromYaml("src/test/resources/nfr.yml"))
}
Java example
import static ru.tinkoff.gatling.javaapi.Assertions.assertionFromYaml;
class test extends Simulation {
setUp(
scn.inject(
atOnceUsers(10)
).protocols(httpProtocol)
).maxDuration(10)
.assertions(assertionFromYaml("src/test/resources/nfr.yml"))
Kotlin example
import ru.tinkoff.gatling.javaapi.Assertions.assertionFromYaml;
class test extends Simulation {
setUp(
scn.inject(
atOnceUsers(10)
).protocols(httpProtocol)
).maxDuration(10)
.assertions(assertionFromYaml("src/test/resources/nfr.yml"))
This extension introduce new syntax (startTransaction
/endTransaction
) for gatling scenario. Transaction is union of
actions, that able to measure summary response time of actions with pauses. It is same as groups, but response time
measuring include pauses, and you may pass endTime manually. That make possible write something like:
exec(Actions.createEntity())
.startTransaction("transaction1")
.doWhile(_ ("i").as[Int] < 10)(
feed(feeder)
.exec(Actions.insertTest())
.pause(2)
.exec(Actions.selectTest)
)
.endTransaction("transaction1")
.exec(Actions.batchTest)
.exec(Actions.selectAfterBatch)
Java example:
exec(Actions.createEntity())
.exec(startTransaction("transaction1"))
.exec(Actions.insertTest())
.pause(2)
.exec(Actions.selectTest)
.exec(endTransaction("transaction1"))
.exec(Actions.batchTest)
.exec(Actions.selectAfterBatch)
Kotlin example:
exec(Actions.createEntity())
.exec(startTransaction("transaction1"))
.exec(Actions.insertTest())
.pause(2)
.exec(Actions.selectTest)
.exec(endTransaction("transaction1"))
.exec(Actions.batchTest)
.exec(Actions.selectAfterBatch)
For use this you need gatling with version greater or equal than 3.6.1 and import this in Scenario and Simulations:
import ru.tinkoff.gatling.transactions.Predef._
Attention!
Your simulation should inherit the class SimulationWithTransactions
instead of Simulation
, then the transaction
mechanism will work correctly.
import io.gatling.core.Predef._
import io.gatling.core.structure.ScenarioBuilder
..
.
import ru.tinkoff.gatling.transactions.Predef._
object DebugScenario {
val scn: ScenarioBuilder = scenario("Debug")
.exec(Actions.createEntity())
.startTransaction("transaction1")
.doWhile(_ ("i").as[Int] < 10)(
feed(feeder)
.exec(Actions.insertTest())
.pause(2)
.exec(Actions.selectTest)
)
.endTransaction("transaction1")
.exec(Actions.batchTest)
.exec(Actions.selectAfterBatch)
}
class DebugTest extends SimulationWithTransactions {
setUp(
DebugScenario.scn.inject(atOnceUsers(1))
).protocols(dataBase)
}
See the examples in the examples/ directory.
You can run these from the sbt console with the commands project example
and then gatling:testOnly ru.tinkoff.load.example.SampleSimulation
.
Ensure that the correct InfluxDB parameters are specified in gatling.conf and influx.conf.
To test your changes use sbt test
.
- Scala version: 2.13.8
- SBT version: 1.6.1
- Gatling version: 3.7.4
- SBT Gatling plugin version: 4.1.2
- SBT CI release plugin version: 1.5.10
- json4s version: 4.0.2
- pureconfig version: 0.17.1
- scalatest version: 3.2.10
- scalacheck version: 1.15.4
- scalamock version: 5.2.0
- generex version: 1.0.2
- jwt-core version: 5.0.0
- scala influxdb client 0.6.3
telegram: @qa_load
gatling docs: https://gatling.io/docs/current/general
We use SemVer for versioning. For the versions available, see the tags on this repository.
-
Maksim Sitnikov - profile module - [email protected]
-
Chepkasov Sergey - influxdb, feeders, config, utils modules - [email protected]
-
Kalyokin Vyacheslav - templates module - [email protected]
-
Akhaltsev Ioann - founder and spiritual guidance - [email protected]
See also the list of contributors who participated in this project.
This project is licensed under the Apache 2.0 License - see the LICENSE file for details
TBD