Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update wpa_supplicant.md #523

Open
wants to merge 4 commits into
base: master
Choose a base branch
from
Open

Update wpa_supplicant.md #523

wants to merge 4 commits into from

Conversation

Piping
Copy link

@Piping Piping commented Oct 24, 2020

add details on how to use wpa_cli and wpa_supplicant based on my reading experience and real life experiment on this doc.

add details on how to use wpa_cli and wpa_supplicant based on my reading experience and real life experiment on this doc.
@Duncaen
Copy link
Member

Duncaen commented Oct 24, 2020

At least enabling the service is redundant, users should refer to the linked page if they don't know how services are enabled, otherwise we would have a copy of ln -s foo bar on every page.
sv start is also not necessary as enabled services are always started unless they have a down file.

@Piping
Copy link
Author

Piping commented Oct 24, 2020

@Duncaen does it mean once the soft link happens the runit will run the service immediately?

As a new user who is referring to the doc during instillation, it is a bit easier to have all information needed in the front, especially for the networking setup. And I found void-installer cannot do this for me (wireless-setup) if I don't enable the wpa_supplicant service first.

Review the document flow and make changes based on review.
@Duncaen
Copy link
Member

Duncaen commented Oct 24, 2020

@Duncaen does it mean once the soft link happens the runit will run the service immediately?

Yes it will start as soon as the service directory is scanned, which should happen within 5 or 14 seconds not sure,
in this period, sv start won't work because the service is not scanned and can't be controlled with sv anyways.

Regarding the new update, this is exactly why this should be documented in the appropriate file, now we would have to duplicate a lot more text over all pages that enable a service, which doesn't really make it better.

@Piping
Copy link
Author

Piping commented Oct 24, 2020

@Duncaen Sure. I think the the time delay need to be mentioned . And using sv start to verify the change will be more deterministic.

@Duncaen
Copy link
Member

Duncaen commented Oct 24, 2020

Yes but the right place would be https://docs.voidlinux.org/config/services/index.html#enabling-services and not every single page that wants you to enable a service, linking to that page should be sufficient for new users who don't know how to enable services.

@Piping
Copy link
Author

Piping commented Oct 24, 2020

A little bit of redundancy is not bad at all. Just referring the general enabling services section does not solve the problem of specific service, i.e I have to look up the correct name of service file/directory. Giving the right command up front would be so much easier. Especially referring to my experience when I cannot get Internet access in void installation and starting to search the Internet, find if I am missing wifi firmware etc. And in void-installer shell, there is not auto-completion of directory and filenames.

@Duncaen
Copy link
Member

Duncaen commented Oct 24, 2020

We've decided that redundancy is just hurting maintainability and reading experience, exactly this example was discussed and is mentioned in the style guide:

Command code-blocks should not be used to describe routine tasks documented elsewhere in this Handbook. For example, when writing documentation for the foo package, do not provide a command code-block stating that one should install it via xbps-install foo. Similarly, do not provide code blocks describing how to enable the foo service.

https://github.com/void-linux/void-docs/blob/master/CONTRIBUTING.md#general

Update the link to enabling service section.
Update the link to enabling service section.
@Piping
Copy link
Author

Piping commented Oct 24, 2020

@Duncaen I don't think it hurts reading experience since it provides whatever information is needed by a reader who is looking for the exact command. But I don't want to spend time on arguing thus removing the ln command.

Copy link
Member

@ericonr ericonr left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Please remember to run the ./check script before pushing the PR, and squash all commits that touch this file (but make a different commit if you touch network/index.md

Comment on lines +8 to +10
Then you need to run `rfkill` to make sure your radio device is not blocked.

Run `rfkill unblock id|type` if it is shown blocked.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Information about rfkill should be added to the general network page, because it applies to wpa_supplicant and IWD, as well as any other daemon we decide to add.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It is already in, but we can perhaps expand it. https://docs.voidlinux.org/config/network/index.html

Comment on lines +19 to +26
```
# wpa_cli -i wlp2s0
(wpa)$ add_network 0
(wpa)$ set_network 0 ssid "wifi_name"
(wpa)$ set_network 0 password abc123
(wpa)$ enable_entwork 0
(wpa)$ quit
```
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Wow, their man page fails to even mention these *_network commands... It would be nice if we could get them added to the upstream man page.

to specify which network interface to use via the `-i` option. Not doing so can result in various `wpa_cli` commands (for example, `scan` and
`scan_results`) not producing the expected output.

Sample Setup Commands:
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

"An example of the procedure to connect to an access point called wifi_name can be seen below:"

(wpa)$ quit
```

## More on wpa_supplicant
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I don't love this title. I think the page could be reorganized to not require it.

@@ -3,7 +3,29 @@
The `wpa_supplicant` package is installed by default on the base system. It
includes utilities to configure wireless interfaces and handle wireless security
protocols. To use wpa_supplicant, you will need to enable [the wpa_supplicant
service](#the-wpa_supplicant-service).
service](../services/index.md#enabling-services).
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@ericonr
Copy link
Member

ericonr commented Oct 29, 2020

A little bit of redundancy is not bad at all. Just referring the general enabling services section does not solve the problem of specific service, i.e I have to look up the correct name of service file/directory.

This could be made clearer, but whenever we say "enable service_name service", the directory is named service_name.

Especially referring to my experience when I cannot get Internet access in void installation and starting to search the Internet, find if I am missing wifi firmware etc.

New images will hopefully include the void-docs package, allowing you to access the documentation even when offline.

And in void-installer shell, there is not auto-completion of directory and filenames.

Is it running dash? We can probably add a note to the installation section about switching to bash before doing anything complicated.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants