We have sort of a mish-mash of "service" and "servicePort" and similar.

We should think about changing ALL of the places that actually relate to a service-port (a specific port of a service, not the whole service) and rename them to some new term that means "service port". E.g. "svcpt" or "spt" or "sport".

See kubernetes/kubernetes@8fffae321d66487f77692

Hi,

first thanks for this awesome software. But I have some trouble and I don't know how to proceed. First I try to find out what else I could provide for this report to be a good bug report.

First, I use the latest version (v1.14.0) from openSUSE build Service. I know, I should compile it by my self to avoid any other causes for this behavior. But for now I still use this package from there

It's not possible to enable addons with --no-kubernetes passed because the addons in their current state are deployed to Kubernetes. So if a user tries to enable an addon with --no-kubernetes output a message notifying them that it's not possible without Kubernetes and exit execution.

Related kubernetes/minikube#13461

Command helm repo list act differently when repository config file don't exists:

$ helm repo list
Error: no repositories to show
$ helm repo list -o json
[]
$ echo $?
0

$ rm ~/.config/helm/repositories.*

$ helm repo list -o json
Error: no repositories to show
$ echo $?
1

In both cases helm should return return empty list of repositories without exiting with erro

Would be good to upgrade the commander dependency in @backstage/cli.

Dependabot tried, but it looks like there's some breaking API changes.

backstage/backstage#10192

Opening up to the community to do as a good first issue! 👍

Requirement - what kind of business use case are you trying to solve?

Actual the windows executables (jaeger-agent.exe, jaeger-collector.exe, ...) don't have a version set. It is hard to identify the current used version and check if the actual used version is outdated.

Probl

1. Describe IN DETAIL the feature/behavior/change you would like to see.

Implement WarmPool.RenderTerraform()

2. Feel free to provide a design supporting your feature request.

/kind feature
/area terraform
/lifecycle frozen

Scenario:

• Single vttablet, connecting to external MySQL (happens to be RDS Aurora, but I doubt it matters).
• Single unsharded keyspace (keyspace3), no vschema, with a single table, containing a single row
• Test program using the Python pymysql driver, using the CLIENT.FOUND_ROWS flag, and updating a single row to it's current key value (i.e. no rows are updated, but a single row

Summary

As a consequence of one of my PRs, a bug was introduced into the master branch of the Argo Workflows codebase. This had instant repercussions (see Github Issue here) for new users that were consuming the tool. Whenever the docs provide a link to manifests, they point to the master branch (see [here](https://github.com/argoproj/ar

The idea originated from this comment, see for full context thanos-io/thanos#4908 (comment)

Basically, we could improve the user experience by providing more information on how the stores are being queried, beyond providing just min / max time or just showing that the min / max time is not available, which does not in fact tell users if that store will be que

Add a network option in creating workload

like rancher‘s Networking tab

** During workload creation, user can customize own hosts and DNS search doma

Is your feature request related to a problem? Please describe.
rovide a method to test/verify if the configuration is correct. before reload the config.
Like nginx -t

Describe the solution you'd like
Simple command.
After run, print the test/verification result.

Describe alternatives you've considered

Additional context

Now the helm is related in build, let's move it to root dir.

kubeedge/build/helm -> kubeedge/helm

The library controller-runtime requires setting a logger (by log.SetLogger()) at the first 30s when the application starts, or it would use the default NullLogSink. We should also call it in testing codes.

When we test with ginkgo, ginkgo provides a helpful GinkgoWriter, which hides the output as default, only prints it when the test failed. We'd better use it to keep our testing output

Spent the morning debugging what I thought was a goroutine leak, but turned out to be just a huge number of goroutines being created in a short period of time. However, it seems a lot of the unit tests do leak goroutines. Most of these might just be issues with tests not cleaning up properly, but it makes testing/debugging real leaks quite difficult. Or in my case, the leaks were a red herring - t

Describe the bug
certain pods are missing "standard" labels which are useful for certain scenarios:

app.kubernetes.io/name
app.kubernetes.io/instance

To Reproduce
Steps to reproduce the behaviour:

1. Deploy longhorn
2. Check labels

default labels are missing on:

all csi-.* pods
engine-image
instance-manager
longhorn-csi-plugin

Expected behavior
app.kubernetes.io