request-benchmark: require --content-type in informer mode - #4365
kubernetes-prow[bot] merged 4 commits into
Conversation
|
/test pull-perf-tests-benchmark-list-informer |
|
/assign @serathius |
| apiVersion := fs.String("api-version", "v1", "apiVersion of the target resource (e.g. v1, apps/v1).") | ||
| resource := fs.String("resource", "secrets", "resource name of the target resource (e.g. pods, deployments).") | ||
| contentType := fs.String("content-type", "proto", "Content type for informer requests. Valid values: [json, proto]") | ||
| contentType := fs.String("content-type", "json", "Content type for informer requests. Valid values: [json, proto]") |
There was a problem hiding this comment.
One ask (I know it's nitpicking), I prefer we didn't have a default. I would prefer that for things like content-type the user needs to explicitly specify so you don't need to read code to confirm what is the default.
There was a problem hiding this comment.
it make sense. i will adjust the pr.
fde7c7b to
de56bac
Compare
|
/test pull-perf-tests-benchmark-list-informer /hold until the CI jobs above pass. |
|
/test pull-perf-tests-benchmark-list-informer |
| // Re-create stopCh so that a second start/gather cycle can Dispose | ||
| // cleanly. Dispose closes stopCh, so reusing the one from a previous | ||
| // cycle (or the constructor) would panic with "close of closed channel". | ||
| e.stopCh = make(chan struct{}) |
There was a problem hiding this comment.
yeah, just testing :) i don't want to merge it.
There was a problem hiding this comment.
Ack, recommend testing locally on kind. It's faster but up to you.
b58dfe4 to
2e46f0d
Compare
|
/test pull-perf-tests-benchmark-list-informer |
1 similar comment
|
/test pull-perf-tests-benchmark-list-informer |
|
testing #4376 /test pull-perf-tests-benchmark-list-informer |
|
testing kubernetes/test-infra#37866 /test pull-perf-tests-benchmark-list-informer |
dc3b0cc to
de56bac
Compare
|
/test pull-perf-tests-benchmark-list-informer |
|
I need to bump the number of pods based on: and |
|
testing with kubernetes/test-infra#37875 /test pull-perf-tests-benchmark-list-informer |
|
/test pull-perf-tests-benchmark-list-informer |
|
testing with kubernetes/test-infra#37880 /test pull-perf-tests-benchmark-list-informer |
|
/test pull-perf-tests-benchmark-list-informer |
f89cc6d to
27460f6
Compare
|
/test pull-perf-tests-benchmark-list-informer |
27460f6 to
d2de0db
Compare
|
/test pull-perf-tests-benchmark-list-informer |
d2de0db to
3f21b07
Compare
|
/test pull-perf-tests-benchmark-list-informer |
3f21b07 to
7a14415
Compare
|
/test pull-perf-tests-benchmark-list-informer |
To understand the difference in the number of requests between the two modes, we would have to download the logs from the client binary and maybe even add some instrumentation. |
|
/hold cancel |
|
/lgtm |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: p0lyn0mial, serathius The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:
Approvers can indicate their approval by writing |
make the content-type default to "json" instead of "proto" for the informer mode, matching the http mode.
the only job that runs informer mode sets the content type explicitly to proto.
this PR allows me to also test kubernetes/test-infra#37854