What would you like to be added:
I would like debug logging to show the heart of the processing of a patch request. For example, showing the inputs and outputs of https://github.com/kubernetes/apiserver/blob/fa3c8ad6d7aff4ba13bf7296583ed2815db1be44/pkg/endpoints/handlers/patch.go#L620 .

Why is this needed:

Netdata is not working with MySQL, only MariaDB. This is because this command SHOW USER_STATISTICS in this file only works in MariaDB.

Problem code is https://github.com/netdata/netdata/blob/bfae683664b488df91940a81760459db2b18983c/collectors/python.d.plugin/mysql/mysql.chart.py

My install is http://walkercodetutorials.codes:19999

I'm using TLS on etcd 3.3.12 and have a mix of etcd proxies and masters. The proxies don't seem to honour the --cipher-suites setting, as it ends up with ECDHE-RSA-AES256-GCM-SHA384 despite --cipher-suites=TLS_ECDHE_RSA_WITH_CHACHA20_POLY1305,TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256 being set.

$ pgrep -laf /usr/bin/etcd
21968 /usr/bin/etcd --listen-client-urls https://0.0.0.0:4000 --adv

Do you want to request a feature or report a bug?

Documentation improvement

In digestAuth documentation, Kubernetes section does not contain any useful example explaining how to put credentials into a file. The given example is same as users(using base

Summary

Currently Kong supports writing configuration in a file, in YAML or JSON.
To provide a more flexible configuration, especially for those running Kong in different environments (dev, staging, production), an option to write configuration in lua can be very useful.

Additional Details

There are many use-cases that come to mind:

• Using environment variables inside the d

Currently at 43.2%.

Sorry in advance but the issue template really does not apply at all to my issue.

Abstract

When using the none driver in a linux environment I get issues with PVCs after restarting. This is due to the /tmp directory being a tmpfs filesystem.

Details

Many mainstream linux distributions adopted systemd as the init system quite a while ago, and under this system, the /tmp director

Move all Dockerfile samples to non-root user

Expected Behaviour

As an OpenFaaS end-user, I want to run all samples on my OpenFaaS cluster, even with the new non-root feature enabled.

Current Behaviour

I had some issues and talked on the slack channel with @hickeyma.
In the end the problem was me, not understanding the helm chart name and the release name.

So I had the following situation:
I didn't know in the beginning that the release name had to be unique within a namespace. I used the helm chart name as component name (copy, content, seo, orders etc.) and the release name initial for

**What kind of request is this (question/bug/enhan

Is your feature request related to a problem? Please describe.
The documentation for this chart specifies that you can give it a logLevel parameter that has a default value of 4. It's not immediately obvious to me what this number means or what other numbers I could give to this parameter.

Describe the solution you'd like
It would be nice to enumerate on the documentation what val

Overview

First off, thanks to all of you who work on K3s. It's such an awesome service. I have a use-case which requires running K3s on Docker. I'm following Using Docker as the Container Runtime. I realized after running the K3s install (see version / arguments below) that I also had to install Doc

Issue Description

Type: bug report

Describe what happened

java.lang.IllegalStateException: failed to req API:/nacos/v1/ns/instance/beat after all servers([minion-register:8848]) tried: failed to req API:minion-register:8848/nacos/v1/ns/instance/beat. code:400 msg: java.lang.IllegalArgumentException: dom name can only have these characters: 0-9a-zA-Z-._:, current: DEFAULT_GROU

Overview of the Issue

docker/test/run.sh doesn't work on macOS.

The script is trying to mount /dev/log that doesn't exist on macOS. We should probably use /var/run/syslog.

Reproduction Steps

Run docker/test/run.sh mysql57 "make build" on macOS.

Operating sy

Some documentation could be improved by adding some options to runsc spec to allow users to specify the command and some other common OCI options. This could also improve the getting started experience for folks who are kicking the tires with runsc.

I'm thinking of the following options:

• Option to specify the command
• Option to specify working directory
• Option to specify network names

Hi, I just upgraded at the latest version and having issues during the new "Waiting for deployments to stabilize" stage. It seems that Skaffold is trying to get infos about resources outside the user's namespace, but I'm not sure. Anyway, it worked perfectly before, with Skaffold v1.5.0.
The only way to make it to work with 1.6.0 version, is turning off the status-check with `--status-check=f

Copied from briandealwis/cram#5: @student-06 asked

It would be great if we can supply auth creds for registry using parameters since it removes the dependency on credential helpers.

My exact use case is: We are setting up a CI system for UI projects, where we need to push docker images (without daemon) to a local registry which has auth.

/kind feature

Persona: Infrastructure Engineer
Control Plane Walkthrough:

• Install Kubeflow with (istio) KfDefs
• Install Kubeflow with dex - pending
• Configure / Get Istio IngressGateway endpoint
• Create profiles (w/ kubectl) #4725
• Create notebooks from user-namespace (w/ kubectl)
• Login (port-forward) into notebook as user
• Create TFJob (w/ kubec

What would you like to be added:

Enable Dynamic Persistent volume Provisioning for ceph rbd provisioner through adding "allowVolumeExpansion" parameter.

Why is this needed:

Dynamic Persistent volume Provisioning is necessary and ceph rbd support this.

https://gravitational.com/teleport/docs/ssh_one_login/#enable-saml-authentication

OneLogin have updated their UI, we should re-run our docs and update images.

I actually saw this in code review but then I noticed the old PNGs were also around 1-2 MB. Probably a little room for improvement if we pass these to online PNG optimizers. It seems like the main issue is that the resolution is simply too large.

Installation method "Method

Description

Add Azure notebook to our SETUP doc.
I tested google colab and Azure notebook to run reco-repo without requiring creating any DSVM or compute by myself, and it works really well with simple tweaks to the notebooks (e.g. for some libs, should install manually).

I think it would be good to add at least Azure notebook to our SETUP doc, where users can easily test out our repo w/o