antiX-25-full-beta2 for public testing

[anticapitalistaForum Admin]

ProwlerGr wrote:

First defect to report
Conky no longer comes up on desktop.
Desktop menu toggle doesn’t work either

Update:
Only icewm variants affected. Definetly related to the upgraded icewm version that just came through

Confirmed

Philosophers have interpreted the world in many ways; the point is to change it.

antiX with runit - leaner and meaner.

[WallonMember]

Bonjour,

I did my first antiX beta 2 “runit” installation on an external SATA SSD drive. It wasn’t easy for me.
As @andfree warned us, I disabled the Trixie backport repository enabled by default on the Live USB stick.
I also enabled the closest and most secure repositories for the Live USB key in my country.
I ran “sudo apt update” directly on the Live USB key.
I did not run “sudo apt full-upgrade” on the Live USB key.
I had to restart the installation several times. I don’t know why it failed because I didn’t get an error message in the log during installation.
Since I wanted to keep the correct repositories from the Live USB stick, I thought I had to ask the installer to keep the changes made on the Live USB stick desktop. Since the installation failed, I did not activate this option during the installation.
During installation, I disabled two services:
– battery management, since I use a tower/desktop PC and don’t have a battery,
– Bluetooth, because my tower/desktop PC doesn’t have a Bluetooth card/chipset.

I checked AranDR for my two screens and set the main screen. Since it is the default “nouveau” driver, I have screen tearing in Firefox ESR when scrolling up and down text quickly in a window.

I answered “Yes” to all the questions for the Antix updates to get all the fixes made by the development team.
As always, by answering “Yes,” I lost the French translations of my menus.
In any case, as Marcelocripe has been reminding us for two years, runit translations are no longer updated for service management (top services and bottom services, for example).

Conky;
a) During installation, Conky displayed “runit.”
b) After the first update, Conky displayed “init.” “runit” no longer existed!
c) After the second update, Conky no longer appears at all.

I noticed in the antiX package installer that it was no longer possible to install “Google Chrome.” I don’t know if this was a decision made by the development team.

It is an installation made on an external SATA SSD drive. The /tmp directory is also not stored in RAM with “runit”. As with “Sysvinit,” there is no tmpfs for “runit.”

wallon@msi:~
$ LANGUAGE=C mount | grep /tmp
wallon@msi:~
$ LANGUAGE=C df -h /tmp
Filesystem      Size  Used Avail Use% Mounted on
/dev/sdc6        67G   12G   52G  18% /
wallon@msi:~
$ 

Best,
Wallon

• This reply was modified 8 months, 3 weeks ago by Wallon.

[anticapitalistaForum Admin]

If you disable trixie backports, you will not get any upgrades of libreoffice, yt-dlp and firmware-linux-nonfree.

@Wallon – read this post

https://www.antixforum.com/forums/topic/antix-25-beta2-upgrades/

Philosophers have interpreted the world in many ways; the point is to change it.

antiX with runit - leaner and meaner.

[Brian MasinickModerator]

Don’t know for sure if this is just my problem or it’s suggestive of something that needs fixing; I still need to go through logs to see if I can discover anything useful about this:

https://www.antixforum.com/forums/topic/what-are-you-here-with-today/page/380/#post-193210

--
Brian Masinick
Alternate "Search B":
This search page

[WallonMember]

anticapitalista wrote:

If you disable trixie backports, you will not get any upgrades of libreoffice, yt-dlp and firmware-linux-nonfree.

@Wallon – read this post

https://www.antixforum.com/forums/topic/antix-25-beta2-upgrades/

OK, I’ll read your message three times.

If I understand correctly, depending on the init used, we won’t have the same applications in the antiX package installer?

[WallonMember]

anticapitalista wrote:

grindstone wrote:

zathura install (32 bit sysV)

ln: failed to create symbolic link ‘/usr/share/runit/sv.now/rc.local/supervise’: Not a directory

Already reported severaltimes.
Here’s the fix.

Rename the file /usr/share/runit/sv.now/rc.local to rc.local1
Create a directory in /usr/share/runit/sv.now/rc.local
Copy the rc.local1 file into that directory and rename it back to rc.local

I also forgot to do what you reminded us about for rc.local.

[anti-apXosMember]

ProwlerGr wrote:

First defect to report
Conky no longer comes up on desktop.
Desktop menu toggle doesn’t work either

Update:
Only icewm variants affected. Definetly related to the upgraded icewm version that just came through

See fix mentioned here: https://www.antixforum.com/forums/topic/icewm-4-is-out/#post-193226

By the way, the default user ~/.icewm/preferences antix sets up from /etc/skel is still based on version 3.3.5, so it’s missing a lot of configs that have been added in recent versions, including the alpha preference. It would be good to update that to 4.0.0 or whatever ICeWM version ends up shipping with Stephen Kapos.

Run

icewm --rewrite-preferences

to write an updated preferences file with all the newer options in it, but everything that was modified from the Icewm defaults kept, too.

"--"

[Brian MasinickModerator]

Brian Masinick wrote:

Don’t know for sure if this is just my problem or it’s suggestive of something that needs fixing; I still need to go through logs to see if I can discover anything useful about this:

https://www.antixforum.com/forums/topic/what-are-you-here-with-today/page/380/#post-193210

I’m running the original live instance of Beta 2 right now and I DO NOT experience this DELAY when I run live.
I’ll try it again on a snapshot and see if there is any difference. Then I may create a couple different INDIVIDUAL init instances other than sysVinit and see if I see this delay or if it’s absent in these experiments. I have not been able to find any hints so far in the logs that I’ve looked at.

Does anyone besides me see a display on the desktop when a particular window manager – IceWM, JWM, Fluxbox, etc. is selected when you’re starting with runit, s6-rc, s6-66 or dinit? Moreover, if you start with anything other than sysVinit, is there a significant delay before the desktop or workspace is available and usable?

I’m NOT seeing this when I run live, so I’m wondering if I created a problem when I installed it or if the installer is adding things that are contributing to what I’m experiencing. Also, if there are recommendations for things I can be examining I’d appreciate that, and if I find anything to share, I’d be happy to do it; I just haven’t found the EVIDENCE to explain this behavior YET.

--
Brian Masinick
Alternate "Search B":
This search page

[anticapitalXMember]

Brian Masinick wrote:

I’m NOT seeing this when I run live, so I’m wondering if I created a problem when I installed it or if the installer is adding things that are contributing to what I’m experiencing.

on one if my installations, during the boot , I had a 10-15 sec delay , the log said it was searching for hibernation uuid or something similar .
then I ran
sudo make-fstab
and the delay was gone .
I hope this solution can help you

^{Let's accept antiX as it is
Pure and Elegant by its nature}

[anticapitalistaForum Admin]

anti-apXos wrote:

ProwlerGr wrote:

First defect to report
Conky no longer comes up on desktop.
Desktop menu toggle doesn’t work either

Update:
Only icewm variants affected. Definetly related to the upgraded icewm version that just came through

See fix mentioned here: https://www.antixforum.com/forums/topic/icewm-4-is-out/#post-193226

By the way, the default user ~/.icewm/preferences antix sets up from /etc/skel is still based on version 3.3.5, so it’s missing a lot of configs that have been added in recent versions, including the alpha preference. It would be good to update that to 4.0.0 or whatever ICeWM version ends up shipping with Stephen Kapos.

Run

icewm --rewrite-preferences

to write an updated preferences file with all the newer options in it, but everything that was modified from the Icewm defaults kept, too.

Sure – where is it?
Sorry I see it now – thanks

• This reply was modified 8 months, 3 weeks ago by anticapitalista. Reason: Read the post better

Philosophers have interpreted the world in many ways; the point is to change it.

antiX with runit - leaner and meaner.

[Brian MasinickModerator]

anticapitalX wrote:

sudo make-fstab

I’ll try this and see if it fixes this issue. Oddly where it’s showing up, even after running, is in erratic network activity. By that I mean that ping will, at one moment, ping reliably on sites that might take 15-25 ms to ping; then just as quickly, it’ll miss a ping, and show 50% ping loss when the sample size is two; do it a few moments later and it’s OKAY. This flatly does not occur with sysVinit in the same setup, nor on any of the other distributions I run here and it doesn’t happen when I run antiX 25 Beta 2 Live from USB either.

I’m going to reboot now and see if this magically fixes everything; it’d sure be nice if that’s the case!

--
Brian Masinick
Alternate "Search B":
This search page

[Brian MasinickModerator]

anticapitalX wrote:

Brian Masinick wrote:

I’m NOT seeing this when I run live, so I’m wondering if I created a problem when I installed it or if the installer is adding things that are contributing to what I’m experiencing.

on one if my installations, during the boot , I had a 10-15 sec delay , the log said it was searching for hibernation uuid or something similar .
then I ran
sudo make-fstab
and the delay was gone .
I hope this solution can help you

We must have something different occurring; I didn’t see how altering a filesystem table would resolve my issue; my issue is NOT hibernation; I’ve not even set myself up with that. My issue is twofold.
1) As the system forks processes and attempts to start the selected window manager, something gets in the way and slows that activity, but it DOES complete. NOTE that this DOES NOT occur with sysVinit, but it DOES occur with runit, which is where I first detected it, and it also happens with dinit, s6-66 and s6-rc.
2) Once the environment is live, it works, but it has something that’s interfering with smooth network operation. My evidence of this is from doing ping requests. I often use ping -c2 to send two packets, and I select network sites that USUALLY respond well; common stuff like google.com, furman.edu (nearby), or nfl.com – the NFL site responded well; a LOT of people are watching today, so that’s a good test.

Unfortunately I have not identified any useful log to get to the underlying cause of what “seems” to be network interference or a rogue process somewhere. I repeat, on the same hardware, even the same distribution running a different init (sysVinit) none of this takes place.

--
Brian Masinick
Alternate "Search B":
This search page

[rejMember]

Hi-

I am testing different beta2 desktops on an external disk, and have made snapshots twice.

They boot to a blinking cursor on my laptop.

Is there a way to fix it from the installed antiX 23, MX, or a snapshot made from beta2 snapshot usb made from installed version on the laptop internally (that does boot successfully)? “Boot repair”?

[ProwlerGrMember]

anti-apXos wrote:

See fix mentioned here: https://www.antixforum.com/forums/topic/icewm-4-is-out/#post-193226

By the way, the default user ~/.icewm/preferences antix sets up from /etc/skel is still based on version 3.3.5, so it’s missing a lot of configs that have been added in recent versions, including the alpha preference. It would be good to update that to 4.0.0 or whatever ICeWM version ends up shipping with Stephen Kapos.

Run

icewm –rewrite-preferences

to write an updated preferences file with all the newer options in it, but everything that was modified from the Icewm defaults kept, too.

Thanks @anti-apXos, your solution worked like a treat.

To sum up the solution in a single liner for users to run:

echo "Alpha=0" >> ~/.icewm/preferences && icewm --rewrite-preferences && icewm -r

[Brian MasinickModerator]

Brian Masinick wrote:

anticapitalX wrote:

Brian Masinick wrote:

I’m NOT seeing this when I run live, so I’m wondering if I created a problem when I installed it or if the installer is adding things that are contributing to what I’m experiencing.

on one if my installations, during the boot , I had a 10-15 sec delay , the log said it was searching for hibernation uuid or something similar .
then I ran
sudo make-fstab
and the delay was gone .
I hope this solution can help you

We must have something different occurring; I didn’t see how altering a filesystem table would resolve my issue; my issue is NOT hibernation; I’ve not even set myself up with that. My issue is twofold.
1) As the system forks processes and attempts to start the selected window manager, something gets in the way and slows that activity, but it DOES complete. NOTE that this DOES NOT occur with sysVinit, but it DOES occur with runit, which is where I first detected it, and it also happens with dinit, s6-66 and s6-rc.
2) Once the environment is live, it works, but it has something that’s interfering with smooth network operation. My evidence of this is from doing ping requests. I often use ping -c2 to send two packets, and I select network sites that USUALLY respond well; common stuff like google.com, furman.edu (nearby), or nfl.com – the NFL site responded well; a LOT of people are watching today, so that’s a good test.

Unfortunately I have not identified any useful log to get to the underlying cause of what “seems” to be network interference or a rogue process somewhere. I repeat, on the same hardware, even the same distribution running a different init (sysVinit) none of this takes place.

Perhaps THIS will help to explain my “frustration”. I just hand timed how long it takes to get into the system straight to the desktop.

With sysVinit, it’s NICE – rough estimate is seven seconds. With the others, more than FOUR times longer, more like 4 1/2 times longer. Just hand estimated with dinit and it was between 32 and 33 seconds. I have NO IDEA what’s holding up the other four init alternatives, but for me to be frustrated it takes a lot. I’ve had no success looking in various logs for any kind of reasonable explanation to this, so it’s difficult for me to provide anything to the team; I’ll keep trying; maybe my head will bump into the answer!

--
Brian Masinick
Alternate "Search B":
This search page

[Author]

Posts