* Add no_proxy settings one might need to use minikube and kubectl behind a proxy.
* add extra line before shell block for formatting
* include suggestions from chenopis
* Fix installing autocompletion documentation
The written solution did not work for me with the error `-bash: __git_ps1: command not found`. Since the auto-completion is already installed you can simply add the Kubernetes autocompletion to it.
* Remove sourcing bash-completion in .bash_profile
* Removed -fs from troubleshooting command
* Add YAML file for configmaps in concepts
* Add new topic for configmaps in concepts
* Add configuration entry for understanding configmaps
* Move Understanding Configmaps to a separate concept file
* Revert "Move Understanding Configmaps to a separate concept file"
This reverts commit 71bd80d7bccd090d0aa1829ae9e4db6cc4dc590c.
* Revert "Add configuration entry for understanding configmaps"
This reverts commit c27fc13f56453dff6e6b8b8f1f9641cbad8ace53.
* Revert "Add new topic for configmaps in concepts"
This reverts commit 3d535f59170a4b5cb290975871f167217484e7ab.
* Revert "Add YAML file for configmaps in concepts"
This reverts commit 8ecabd6d106647631819441be8bf7e834f0dbb54.
Addresses the work described at #4134.
- Remove /docs/samples page: This page mostly had links to examples on
kubernetes/kubernetes (examples dir) which is now at kubernetes/examples
repository. It was linking to README.md files for tutorials. Now removing
this page as we are moving tutorials from kubernetes/examples to here:
- Create redirects from /docs/samples/ to /docs/tutorials/ (both JS and
redirect_from).
- Deprecate "MEAN Stack" tutorial: This tutorial overlaps quite a bit with
existing "maintained" examples and has been out-of-date for a while.
Creating an external redirect to its original and more up-to-date location:
https://medium.com/google-cloud/running-a-mean-stack-on-google-cloud-platform-with-app-engine-and-mongolab-4bbd2040ea75
- Import README.md files of maintained examples (guestbook, wordpress+mysql and
cassandra) to this repository and serve them under /docs/tutorial.
- Update /docs/tutorials ToC:
- Link to imported tutorials.
- Fix missing articles on sidebar (_data/tutorials.yml) and make the tutorials
listed on tutorials/index.md available on the sidebar (closes#4007).
Signed-off-by: Ahmet Alp Balkan <ahmetb@google.com>
* Update installation instructions with snap
Update installation instructions with snap - the package manager for Ubuntu and other Linux distributions.
* fix md formatting
* fix shell formating
* fix shell block
* fix shell for Check the kubectl configuration
* switch to full width shell
* Update scratch.md: include download binary tarball
As `kubernetes-server-linux-amd64.tar.gz` is not provided with the tarball now, an extra step is required.
* Updated scratch.md to follow review comments
* rename kargo to kubespray in getting started docs
* remove old kargo guide, update link to kubespray doc
* update setup.yml with new kubespray pointer
If a default storage class exists, it is going to create a new persistent volume to bind to that claim instead of using our example one, which will make the tutorial fail. In particular, this happens on minikube.
Specifying an empty storageClassName will prevent the storage class to be used, and force our volume to be selected.
Using `--no-check-certificate` is, frankly, dangerous and we shouldn't be recommending that users use it. Not without a good reason, anyway.
If a user's system is sufficiently broken that they are getting certificate validation problems during this step, I fail to see how it is our problem to tell the user to overcome that. Let's focus on installing Kubernetes here.
`sudo su -` is redundant, as you are asking the system to elevate you after you have already elevated.
If what you actually want is to get a login environment, the `sudo -i` flag is designed exactly for that.
Docker documentation (https://docs.docker.com/engine/reference/run/) reads:
"To modify the proportion from the default of 1024, use the -c or --cpu-shares flag to set the weighting to 2 or higher. If 0 is set, the system will ignore the value and use the default of 1024."
So the Kubernetes documentation gives the impression that if requests.cpu is zero, Docker would end up applying a value of 1024, which isn't the case.
This change makes the explanation more precise, hopefully without making it much more difficult to read.
Christoph Blecker