github/kubevela
1
0
Fork 0
mirror of https://github.com/kubevela/kubevela.git synced 2026-02-14 18:10:21 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix: typo in load-test readme (#4560)

Browse Source 
Signed-off-by: Somefive <yd219913@alibaba-inc.com>
This commit is contained in:
Somefive
2022-08-04 17:27:41 +08:00
committed by GitHub
parent 9fae4df34e
commit 6fbbdc97ca
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
hack/load-test/README.md
View File

@@ -1,4 +1,4 @@
# Pressure Test Guide
# Load Test Guide
## Files
@@ -13,4 +13,4 @@ The `dashboard.json` is a grafana dashboard file that can track several groups o
3. **[Optional] Prepare Kubemark Environments.** The `tolerate-hollow-node` trait need to be installed if you would like to make load test on hollow nodes mocked by `kubemark`. Another issue for using `kubemark` is that you need to ensure pods created by pressure test applications would be assigned to hollow nodes while other pods like KubeVela controller must not be assigned on these hollow nodes. You can achieve that by adding taints to nodes and adding nodeSelector or toleration to pods.
4. **Prepare Monitoring Tools.** Install Grafana, Loki, Prometheus in your Kubernetes which can help you watch load test progress and discover problems. If you have `vela` command line tool installed, you can run `vela addon enable observability` to enable it.
5. **Run Pressure Test.** Start load test by rendering application templates with different IDs to generate application instances and apply them to Kubernetes at a desired creation speed. Wait for a while (could be hours) and delete them. This is standard progress of the pressure test. More mutating actions could be injected.
6. **Check Result.** You can upload the grafana dashboard file to the Grafana website exposed from your cluster. Then you can check the result of the load test.
6. **Check Result.** You can upload the grafana dashboard file to the Grafana website exposed from your cluster. Then you can check the result of the load test.