Commit Graph

60 Commits (183e6cc1902c2088a40e60cf4138bd829d4ad80f)

Author SHA1 Message Date
Shyam Jeedigunta d477bd904c Leave slack for addon pods in node's capacity 2019-01-22 20:26:02 -08:00
Maciej Borsz 25c2440996 Add node notready/unreachable tolerations to load/density pods. 2018-12-14 12:16:51 +01:00
wojtekt 68bae382b1 Fix density test 2018-11-14 10:17:40 +01:00
Mike Danese 7613482623 scale test service account token projection in kubemark 2018-10-26 09:57:56 -07:00
Chris O'Haver 402266afde add dns pod memory constraint 2018-10-18 10:35:30 -04:00
Guoliang Wang c2622dd9d8 Replace Parallelize with function ParallelizeUntil and formally deprecate the Parallelize 2018-10-05 17:56:56 +08:00
Shyam Jeedigunta 203664933d Add etcd DB size monitoring in density test 2018-08-30 14:40:59 +02:00
Maciej Borsz 2632a6a619 Scalability tests: Increase sample size for pod startup latency measurement. 2018-08-03 12:24:43 +02:00
tanshanshan af0fc11ded error in return value should be tested 2018-07-30 15:17:32 +08:00
Maciej Borsz 9df9c2e506 Modify resource constraints for master components.
The growth is caused by increasing --kube-api-qps in kube-scheduler.
2018-07-13 16:46:57 +02:00
Kubernetes Submit Queue a8777c26fa
Merge pull request #64695 from krzysied/etcd_metrics
Automatic merge from submit-queue (batch tested with PRs 64695, 65982, 65908). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.

Collecting etcd metrics

**What this PR does / why we need it**:
Adding etcd metrics to performance test log.

**Which issue(s) this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close the issue(s) when PR gets merged)*:
ref #64030

**Special notes for your reviewer**:

**Release note**:

```release-note
NONE
```
2018-07-10 08:55:03 -07:00
Shyam Jeedigunta 0c787703f5 Introduce scheduler CPU/Memory profile-gathering in density test 2018-06-22 12:12:05 +02:00
Shyam Jeedigunta 457548ef7d Refactor profile-gatherer to work across all master components 2018-06-22 12:11:56 +02:00
Kubernetes Submit Queue 65a5e68147
Merge pull request #64521 from shyamjvs/compute-scheduler-throughput-avg
Automatic merge from submit-queue. If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.

Compute avg and quantiles of scheduler throughput in density test

Based on my comment here - https://github.com/kubernetes/kubernetes/pull/64266#issuecomment-393189953

/sig scheduling
/kind cleanup
/priority important-soon
/milestone v1.11
/cc @wojtek-t 

```release-note
NONE
```
2018-06-13 14:23:51 -07:00
Shyam Jeedigunta 979a8d73e1 Compute avg and quantiles of scheduler throughput in density test 2018-06-12 18:40:52 +02:00
Krzysztof Siedlecki aa022310a4 Collecting etcd metrics 2018-06-04 16:23:08 +02:00
Kubernetes Submit Queue a2d8636559
Merge pull request #64316 from krzysied/scheduling_latency_metric
Automatic merge from submit-queue (batch tested with PRs 63328, 64316, 64444, 64449, 64453). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.

Fixing scheduling latency metrics

**What this PR does / why we need it**:
Allows to measure and to display scheduling latency metrics during tests. Provides new functionality of resetting scheduler latency metrics.

**Which issue(s) this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close the issue(s) when PR gets merged)*:
Fixes #63493

**Special notes for your reviewer**:
E2eSchedulingLatency, SchedulingAlgorithmLatency, BindingLatency are now available 
as subtypes of OperationLatency.

**Release note**:

```release-note
NONE
```
2018-05-30 08:42:16 -07:00
Krzysztof Siedlecki 0e833bfc83 Fixing scheduling latency metrics 2018-05-30 11:20:12 +02:00
Shyam Jeedigunta 9ea33c604c Fix bug with scheduler throughput variable pass-by-value 2018-05-28 13:23:37 +02:00
Kubernetes Submit Queue 5f578f3385
Merge pull request #63979 from soltysh/drop_reapers
Automatic merge from submit-queue (batch tested with PRs 63859, 63979). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.

Drop reapers

/assign @deads2k @juanvallejo 

**Release note**:
```release-note
kubectl delete does not use reapers for removing objects anymore, but relies on server-side GC entirely
```
2018-05-26 00:32:11 -07:00
Maciej Szulik 383872615d
Remove kubectl reapers 2018-05-25 22:18:05 +02:00
Shyam Jeedigunta f363f549c0 Measure scheduler throughput in density test 2018-05-25 14:49:11 +02:00
Kubernetes Submit Queue 7eb88f11d2
Merge pull request #59727 from wgliang/master.time
Automatic merge from submit-queue. If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.

should use time.Since instead of time.Now().Sub

**What this PR does / why we need it**:
should use time.Since instead of time.Now().Sub

**Special notes for your reviewer**:
2018-05-10 20:29:40 -07:00
wojtekt 4cb99318d2 Fix PodStore to wait for being initialized 2018-04-12 15:38:11 +02:00
Wang Guoliang 89669283fe should use time.Since instead of time.Now().Sub 2018-04-10 12:05:51 +08:00
Manjunath A Kumatagi 1bb810e749 Use pause manifest image 2018-04-06 11:00:50 +05:30
Shyam Jeedigunta a44e335940 Don't quit without printing API latencies in density test if it failed 2018-04-03 17:37:49 +02:00
Kubernetes Submit Queue 74c2d3d11a
Merge pull request #61715 from shyamjvs/increase-density-cm-threshold
Automatic merge from submit-queue (batch tested with PRs 60499, 61715, 61688, 61300, 58787). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.

Increase cpu/mem thresholds for c-m in density test

Follows https://github.com/kubernetes/kubernetes/issues/61190#issuecomment-376159552

/cc @wojtek-t 

```release-note
NONE
```
2018-03-27 04:08:06 -07:00
p0lyn0mial 1f1d24005a removes custom scalers from kubectl 2018-03-26 19:33:42 +02:00
Shyam Jeedigunta c27531f2a1 Increase cpu/mem thresholds for c-m in density test 2018-03-26 15:36:50 +02:00
Shyam Jeedigunta 0f0c754eb4 Get rid of duplicate VerifyPodStartupLatency util in node density tests 2018-03-21 16:58:31 +01:00
Shyam Jeedigunta b0dd166fa3 Capture different parts of pod-startup latency as metrics 2018-03-21 16:58:25 +01:00
Shyam JVS b43b621690
Increase apiserver mem-threshold in density test 2018-03-13 16:47:14 +01:00
Shyam Jeedigunta e6c60a7465 Add retries to resource deletions in testing framework 2018-03-01 15:18:46 +01:00
Ryan Hitchman e04b91facf Remove unused variables (only assigned to) from test code.
This is revealed by the go/types package, which is stricter than
the Go compiler about unused variables. See also: golang/go#8560
2018-02-27 13:45:31 -08:00
Shyam Jeedigunta 9f84841d8a Add smart retries to resource creations in testing framework 2018-02-22 21:35:06 +01:00
Marek Grabowski 77a1268fed Add quotas to density and load tests 2018-02-16 09:53:26 +00:00
Shyam Jeedigunta db0a3721f0
Fix bug with profile-gathering waitgroup in scale tests 2018-02-08 15:58:51 +01:00
Shyam Jeedigunta c3db0828f9
Add apiserver profiling to our scalability tests 2018-02-08 11:09:31 +01:00
p0lyn0mial dd9de90b0a the changes introduced in this commit plumbs in the generic scaler into kubectl.
note that we don't change the behaviour of kubectl.
For example it won't scale new resources. That's the end goal.
The first step is to retrofit existing code to use the generic scaler.
2018-01-12 09:21:18 +01:00
Kubernetes Submit Queue c4bddea55d
Merge pull request #56202 from porridge/typo-density
Automatic merge from submit-queue (batch tested with PRs 55557, 55504, 56269, 55604, 56202). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.

Fix typo.

**Release note**:
```release-note
NONE
```
2017-12-14 13:54:49 -08:00
Marcin Owsiany ceda4c31f9 Bump CPU requirements of L7 LB controller. 2017-11-23 11:21:23 +01:00
Marcin Owsiany b31cffd76d Fix typo. 2017-11-22 11:15:59 +01:00
Kubernetes Submit Queue fe599c7dcf
Merge pull request #54992 from porridge/perf-timing
Automatic merge from submit-queue. If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>.

Add performance test phase timing export.

**What this PR does / why we need it**:

First step totwards allowing us to get a quick overview of test length
via perf-dash.k8s.io.

**Release note**:
```release-note
NONE
```

@kubernetes/sig-scalability-feature-requests
2017-11-11 01:39:30 -08:00
Marcin Owsiany 96089e0b79 Add performance test phase timing export. 2017-11-09 17:12:15 +01:00
Dr. Stefan Schimanski 012b085ac8 pkg/apis/core: mechanical import fixes in dependencies 2017-11-09 12:14:08 +01:00
Kevin 4c8539cece use core client with explicit version globally 2017-10-27 15:48:32 +08:00
Shyam Jeedigunta 10254826ec Increase l7-lb-controller's memory check threshold in density test 2017-10-17 09:28:29 +02:00
Shyam Jeedigunta 419bbd26fc Retry if possible while creating latency pods in density test 2017-09-19 17:40:57 +02:00
Kubernetes Submit Queue 3c8fb4b90f Merge pull request #52426 from shyamjvs/dont-crash-on-missing-data
Automatic merge from submit-queue

Don't crash density test on missing a single measurement

We failed our last run due to this (https://k8s-gubernator.appspot.com/build/kubernetes-jenkins/logs/ci-kubernetes-e2e-gce-scale-performance/33) and didn't have pod-startup latency recorded at all.
2017-09-14 05:09:46 -07:00