If not, using `go test -count=n` would make them pile up and ultimately
get to the limit of open files:
apiserver_test.go:984: 17: unexpected error: Get http://127.0.0.1:44131/apis/test.group/version/simple: dial tcp 127.0.0.1:44131: socket: too many open files
Steps to reproduce (no longer fails):
godep go test -short -run '^$' -o test .
./test -test.run '^TestList$' -test.count 100
Note that this might not fail if your `ulimit -n` is not low enough.
This change compare the timestamps in pod status to accurately determine
whether the initial delay has passed when the pod (container) becomes ready.
Also increase the test timeout to account for the parallel testing environment.
Job configs are stored in YAML files in hack/jenkins/job-configs in the
format used by Jenkins Job Builder. A docker container running on
Jenkins fetches job configs every fifteen minutes and updates Jenkins.
This change refactors the code of preparing kube-system manifests
for trusty based cluster. The manifests used by nodes do not contain
salt configuration, so we can simply copy them from the directory
cluster/saltbase/salt, make a tarball, and upload to Google Storage.
I'm moving/deleting some tests to make an upcoming RESTClient refactor
PR cleaner.
- TestDoRequestBearer is redundant to
pkg/client/transport.TestBearerAuthRoundTripper
- Added a case to pkg/client/transport.TestBasicAuthRoundTripper to
cover what TestDoRequestWithoutPassword was testing
- Moved TestSetsCodec, TestRESTClientRequires,
TestValidateHostParameter to helper_test.go since they were really
testing helper functions and not the RESTClient directly.
- Modified TestValidateHostParameter during the moved to just use
DefaultServerURL instead of creating a client.
The original scale function takes around 800ns/op with more
than 10 allocations. It significantly slow down scheduler
and other components that heavily relys on resource pkg.
For more information see #18126.
This pull request tries to optimize scale function. It takes
two approach:
1. when the value is small, only use normal math ops.
2. when the value is large, use math.Big with buffer pool.
The final result is:
BenchmarkScaledValueSmall-4 20000000 66.9 ns/op 0 B/op 0 allocs/op
BenchmarkScaledValueLarge-4 2000000 711 ns/op 48 B/op 1 allocs/op
I also run the scheduler benchmark again. It doubles the throughput of
scheduler for 1000 nodes case.
- Don't indent top-level lists
- Use 2 space list item indentation, unless using multi-line items, then use 4 space indents
- Use underline for page title
- Auto-generate table of contents
- Start with lvl 2 headers (lvl 1 headers look too much like the page title)
k8s-merge-robot