ols-load-generator

README

ols-load-generator

Load generator tool for openshift lightspeed service (OLS).

Prerequisites

Running on openshift cluster

OLS deployed on openshift cluster. Please refer to the instructions here.

Running on local machine (Optional)

A running instance of OLS (to test). Please refer to the instructions here.

Installation

make build; make install

NOTE: You might want to add sudo to the install command as it involves creating ols-load-generator binary in your $PATH.

If running on openshift simply run make which will build and push the image to specified image registry.

Usage

Usage on openshift platform

Once we have OLS deployed and running on openshift cluster, in order to trigger the load test follow the below steps.

Create a secret with your cluster kubeconfig

oc create secret generic kubeconfig-secret --from-file=kubeconfig=<YOUR-KUBECONFIG-PATH> -n ols-load-generator

Deploy config/ols-load-generator.yaml replaced with your corresponding envs values onto the cluster

Envs

• OLS_TEST_HOST - String indicating OLS endpoint to perform load testing.
• OLS_TEST_UUID(Optional) - String specifying an unique ID. Will be helpful while comparing two runs or while looking at a specific run results.
• OLS_TEST_AUTH_TOKEN - OLS auth token string.
• OLS_TEST_DURATION - Load testing duration on each API endpoint.
• OLS_TEST_WORKERS - Amount of parallel workers to trigger load on OLS. This will basically help us send requests in parallel.
• OLS_TEST_METRIC_STEP - Step size for the cluster prometheus metrics to be captured.
• OLS_TEST_PROFILES - List of metric profiles that contain queries to be executed on prometheus.
• OLS_TEST_ES_HOST(Optional) - Elastic search host url. If not specified metrics will be indexed locally.
• OLS_TEST_ES_INDEX(Optional) - Elastic search index name to store the data. If not specified metrics will be indexed locally.
• OLS_TEST_QUERY_ONLY(Optional) - Flag to enable load tests only on /v1/query and /v1/streaming_query endpoints.

Example Usage

oc apply -f ~/config/ols-load-generator.yaml

Once applied it will create a job in the specified namespace and will start running the tests with above mentioned values. We can tail the logs in order to look at the benchmark results.

Usage on Local Machine

NAME:
   ols-load-generator - A command-line tool to load test openshift lightspeed service (OLS).

USAGE:
   ols-load-generator [global options] command [command options] [arguments...]

DESCRIPTION:
   A command-line tool to load test openshift lightspeed service (OLS).

COMMANDS:
   attack   ols-load-generator attack
   index    ols-load-generator index
   help, h  Shows a list of commands or help for one command

GLOBAL OPTIONS:
   -D          print debugging logs (default: false)
   -W          quieter log output (default: false)
   --help, -h  show help

attack sub-command

Subcommand to trigger the load test on openshift lightspeed service.

NAME:
   ols-load-generator attack - ols-load-generator attack

USAGE:
   ols-load-generator attack [command options] [arguments...]

DESCRIPTION:
   perform attack on ols endpoints

OPTIONS:
   --host value        --host localhost:6060 (default: "http://localhost:6060") [$OLS_TEST_HOST]
   --authtoken value   --authtoken authtoken [$OLS_TEST_AUTH_TOKEN]
   --uuid value        --uuid f519d9b2-aa62-44ab-9ce8-4156b712f6d2 (default: "76d8f64d-2bf3-49ac-82c3-22011ddc2284") [$OLS_TEST_UUID]
   --duration value    --duration 1m (default: 1m0s) [$OLS_TEST_DURATION]
   --workers value     --workers 10 (default: 10) [$OLS_TEST_WORKERS]
   --eshost value      --eshost eshosturl [$OLS_TEST_ES_HOST]
   --esindex value     --esindex esindex [$OLS_TEST_ES_INDEX]
   --metricstep value  --metricstep 30 (default: 30) [$OLS_TEST_METRIC_STEP]
   --profiles value    --profiles metrics.yaml,metrics-report.yaml (default: "attacker/assets/profiles/metrics-report.yaml,attacker/assets/profiles/metrics-timeseries.yaml") [$OLS_TEST_PROFILES]
   --queryonly         --query (default: false) [$OLS_TEST_QUERY_ONLY]
   --help, -h          show help

Example Usage

export KUBECONFIG=<your-kubeconfig-path>
ols-load-generator attack --host https://127.0.0.1:9001 --uuid random-uuid --authtoken 'auth-token' --duration 1m --workers 10

index sub-command

Subcommand to scrape and index cluster prometheus metrics within a given time range.

NAME:
   ols-load-generator index - ols-load-generator index

USAGE:
   ols-load-generator index [command options] [arguments...]

DESCRIPTION:
   Indexes metrics within given timerange

OPTIONS:
   --uuid value        --uuid f519d9b2-aa62-44ab-9ce8-4156b712f6d2 (default: "3c2e0d71-07f6-4bfa-b604-61a3243bbec7") [$OLS_TEST_UUID]
   --eshost value      --eshost eshosturl [$OLS_TEST_ES_HOST]
   --esindex value     --esindex esindex [$OLS_TEST_ES_INDEX]
   --metricstep value  --metricstep 30 (default: 30) [$OLS_TEST_METRIC_STEP]
   --start value       --start 1720410990 (default: 1720482385) [$OLS_TEST_START]
   --end value         --end 1720470990 (default: 1720485985) [$OLS_TEST_END]
   --profiles value    --profiles metrics.yaml,metrics-report.yaml (default: "attacker/assets/profiles/metrics-report.yaml,attacker/assets/profiles/metrics-timeseries.yaml") [$OLS_TEST_PROFILES]
   --queryonly         --query (default: false) [$OLS_TEST_QUERY_ONLY]
   --help, -h          show help

Example Usage

export KUBECONFIG=<your-kubeconfig-path>
ols-load-generator index --profiles "metrics.yaml" --start 1720410990 --end 1720410990