antiX-19-b1-full (64 and 32 bit) available

[Member skidoo]

liveboot help menu
F1 }} Main Menu boot options

This page does not yet mention the splasht=somevalue(s) options

In addition to mentioning splasht there, I would welcome appending a “Debugging” section to the bottom of that help page.

(or, if per-page character length is a limiting factor, a link at the bottom of “Main Menu boot options” leading to a separate “Debugging” page)
.

.

excerpted from docs-antiX-17/FAQ/boot-params.html

Debugging

Sometimes booting problems can be tracked down to a missing driver or a driver that is not getting loaded. These two boot parameters can be helpful in tracking down such problems.

bp=N Set one or more breakpoints in the live-init script separated by commas. Use bp=? to get a list of breakpoints to choose from. The meaning of each may change across different versions so use bp=? be sure.

breakpoint 1 before welcome
breakpoint 2 before coldplug
breakpoint 3 before looking for linuxfs file
breakpoint 4 after mounting boot device
breakpoint 5 after mounting persistence device
breakpoint 6 after mounting aufs
breakpoint 7 before prepare switch_root
breakpoint 8 before running init.d scripts
breakpoint 9 right before starting init

breakpoint F After fsck
breakpoint F Before fsck
breakpoint e On possibly fatal error
breakpoint f in frugal install
breakpoint i after init= chroot
breakpoint m before makefs
breakpoint r After resize homefs rsync
breakpoint r Before resize homefs rsync
breakpoint tr After copy to ram
breakpoint u final umount
breakpoint v before check VID
breakpoint x before select device

verb=M
Control how much gets printed out during the early boot process. The default level is verb=5. Lower numbers print less, higher numbers print more. Usually 6 or 7 will show you more info without being too obnoxious. The full output is available in the /var/log/live/bootloader.log file. A partial version of the log file is available at /init.log from within most breakpoints.

[Member Sadshark [p.n.2001]]

What is the name for the new version gonna be?

Running antinataliX GNU/Linux 19.9 on my Compaq Mini 110c-1100
On my desktop running Kubuntu 18.04 LTS
The antinatalist antifascist childfree vegan mini-manifesto
Bertolt Brecht's beautiful poem, the Unconquerable Inscription.

[Forum Admin BitJam]

Sadshark [p.n.2001] wrote:

What is the name for the new version gonna be?

The code name for the alphas and for beta-1 has been Marielle Franco.

Context is worth 80 IQ points -- Alan Kay

[Member Sadshark [p.n.2001]]

BitJam wrote:

Sadshark [p.n.2001] wrote:

What is the name for the new version gonna be?

The code name for the alphas and for beta-1 has been Marielle Franco.

Okay, I am satisfied with that name! Lol. Keep up the good work devs. Just don’t name the next one Antix 20 Alex Jones or anything like that, yeah? I wouldn’t be able to live with myself.

Running antinataliX GNU/Linux 19.9 on my Compaq Mini 110c-1100
On my desktop running Kubuntu 18.04 LTS
The antinatalist antifascist childfree vegan mini-manifesto
Bertolt Brecht's beautiful poem, the Unconquerable Inscription.

[Member wasp]

I have to install Antix on an old Asus, is the beta stable enough to upgrade it to the final with apt dist upgrade?
Thanks

[Member eugen-b]

BitJam wrote:

eugen-b wrote:

It would be great to implement a more normal copy and paste in urxvt like in Manjaro i3 edition for example, I will look for the config and post a link. https://gitlab.manjaro.org/profiles-and-settings/desktop-settings/blob/master/community/i3/skel/.Xresources

We already have a mechanism for tweaking urxvt called urxvt-style. Use “urxvt-style –help” to get instructions.

The idea is to make it easy to play with different settings like:
urxvt-style tabbed medium size16 zenburn transbg
Once you find your sweet-spot then append the “save” command:
urxvt-style tabbed medium size16 zenburn transbg save
and those settings will be saved in the ~/.Xresources file.

It would be easy to add other options to this system. Or if they are overwhelmingly better than the default perhaps we could add them to ~/.Xresources directly.

I now tried urxvt-style and it works great, I prefer the zenburn colour scheme to the default one in the distro I mentioned.
* But it would be great if the new user wouldn’t start with a white urxvt with small font.
* Which means to get the .Xresources file reliably read at startup without the need of urxvt-style reset command.
* IMO, the options for clipboard are very useful for new users. (They might be undesirable for longtime urxvt users who know how to copy and paste without them.)

! Normal copy-paste keybindings without perls
URxvt.iso14755:                   false
URxvt.keysym.Shift-Control-V:     eval:paste_clipboard
URxvt.keysym.Shift-Control-C:     eval:selection_to_clipboard

UPDATE: @BitJam, I checked that urxvt starts in antix style (black background) on all sessions except herbstluftwm. So the issue with .Xresources is only in hlwm, where you need to run urxvt-style reset or xrdb merge .Xresources on each startup. So it would make sense to put one of these commands into ~/.config/herbstluftwm/autostart as a workaround, or find out if there is something missing in SLiM or antix-sessions for hlwm session.

UPDATE 2: I found the conditions to reproduce it (select herbstluftwm in the boot menu query) and filed a bug report: https://bugs.mxlinux.org/show_bug.cgi?id=115
If I log in to herbstluftwm via SLiM the terminal urxvt is themed according to .Xresources, but I get a session election window which is quite useless at that place, see the bug report for details, no need to flood the topic here more than that.

• This reply was modified 1 year, 1 month ago by eugen-b.
• This reply was modified 1 year, 1 month ago by eugen-b.

[Moderator fatmac]

wasp wrote:

I have to install Antix on an old Asus, is the beta stable enough to upgrade it to the final with apt dist upgrade?
Thanks

I have it installed on an Acer AO532 netbook, this is a second installation, as there seems to be a conflict between ceni & connman, if you try to use ceni, it seems to block access somehow, (this is being looked into), but once connected with connman, seems to be stable.

In my past experience with beta versions, very little gets changed before final releases, so upgrading with each beta release usually works to bring it up to release status.

This being beta1, & only recently released, I have only put it onto my netbook to test for just now, (17.4 & MX are on my other computers).

The choice to use it is yours to make. 😉

Linux (& BSD) since 1999

[Member eugen-b]

Is it intentional that I need to tap with three fingers to get the Fluxbox main menu? One finger tap doesn’t give anything. Happens in all Fluxbox sessions.

@wasp, considering kinds of possible misconfoguration like this one I would wait with a productive installation. Or you will need to fix those in $HOME yourself, or copy the updated configuration from /etc/skel to $HOME (but backup your own changes first), or antiX might have an automatic solution which I don’t know about. But the more people test install betas and RCs, the more they will report feedback, the more effectivly the bugs get fixed.

• This reply was modified 1 year, 1 month ago by eugen-b.
• This reply was modified 1 year, 1 month ago by eugen-b.

[Forum Admin Dave]

eugen-b wrote:

UPDATE: @bitjam, I checked that urxvt starts in antix style (black background) on all sessions except herbstluftwm. So the issue with .Xresources is only in hlwm, where you need to run urxvt-style reset or xrdb merge .Xresources on each startup. So it would make sense to put one of these commands into ~/.config/herbstluftwm/autostart as a workaround, or find out if there is something missing in SLiM or antix-sessions for hlwm session.

That is interesting as I had inserted the xrdb merge command into desktop-session for a while now. I have no idea at the moment why herbstluftwm would be different at the moment…. maybe it does a reset on startup or something.

Computers are like air conditioners. They work fine until you start opening Windows. ~Author Unknown

[Member eugen-b]

It happens when you select herbstluftwm in the boot menu. (Rarely any user would do that.) I filed a bug report on this herbstluftwm session issue https://bugs.mxlinux.org/show_bug.cgi?id=115

[Member Xecure]

eugen-b wrote:

Is it intentional that I need to tap with three fingers to get the Fluxbox main menu? One finger tap doesn’t give anything. Happens in all Fluxbox sessions.

If taping with tree fingers is equivalent to right click, then yes. That is normal with fluxbox (it is like openbox: right clicking opens the menu).

@Devs

For now, all the different options I have tested on live-USB with persistence work as expected.
With boot options, I still haven’t tested what “wicd” and “nowicd” options do. Will they now enable/disable Connman instead of WICD (no longer installed) so that ceni works properly, or are they to be changed for “connman” and “noconnman” in the future?

WIfi not working with Connman in first start
For those having trouble with Connman and Wifi, on my Acer laptop it was fixed after rebooting. What I did:
0. I experienced that, after a fresh live-USB, Connman had wifi disabled by default.
1. I went to Control Center > Network > Wifi Connect (Connman) and enable Wifi (change toggle to blue). No networks were found (it looked like it was not working).
2. I rebooted. On the next session, Connman could detect and connect to wireless networks (all working perfectly).

Tsplash experience
No fails from tsplash for now. On laptop perfect. I really like it very much.
First time booting with the liveUSB on my tablet (which is slow but with an enormous screen resolution), in the step after loading to RAM, I think it displays “Start Init”, the screen went totally black. I waited for about 20 seconds and decided to see what was going on.
I hit ALT+F1 (as the tsplash screen recommended before it went black to see more info). The terminal was there (so all well) and the last command displayed said “waiting for /dev to fully populate”. I returned to the black screen with ALT+F10 and after a few seconds, the tsplash screen returned (I believe that with a change of resolution). So everything was good.
The second time booting from live USB on my tablet was faster and the black screen and resolution change only lasted a second.

Fantastic work, guys!

[Member BobC]

On my Dell D620 laptop, I don’t have any problems with the splash f or va.

When booted from flashdrive I get no WiFi via Connman-gtk. I did enable it. I also tried flipping the WiFi switch off, wait 10 seconds and back on, but my router doesn’t appear. I normally install to hard drive, but was testing if it would “just work” from flashdrive or installed, which would be an optimal solution.

[Member eugen-b]

It boots to a bautiful 1922×1080 resolution on my VIA VX800 box. (Some of you might remember it was always a pain on this machine.)

$ inxi -Fxxxz
System:    Host: antix1 Kernel: 4.9.176-antix.1-486-smp i686 bits: 32 compiler: gcc 
           v: 8.3.0 Desktop: IceWM 1.5.5+git20190610 dm: SLiM 1.3.6 
           Distro: antiX-19.b1_386-full Marielle Franco 12 June 2019 
           base: Debian GNU/Linux 10 (buster) 
Machine:   Type: Desktop Mobo: VIA model: VX800 v: 1.0 serial: <filter> BIOS: Phoenix 
           v: 6.00 PG date: 09/15/2010 
CPU:       Topology: Single Core model: VIA Eden bits: 32 type: UP arch: C7 
           L2 cache: 128 KiB 
           flags: nx pae sse sse2 sse3 bogomips: 797 
           Speed: 400 MHz min/max: 400/1000 MHz Core speed (MHz): 1: 1000 
Graphics:  Device-1: VIA VX800/VX820 Chrome 9 HC3 Integrated Graphics driver: N/A 
           bus ID: 00:01.0 chip ID: 1106:1122 
           Display: x11 server: X.Org 1.20.4 driver: openchrome 
           unloaded: fbdev,modesetting,vesa resolution: 1920x1080~60Hz 
           OpenGL: renderer: llvmpipe (LLVM 7.0 128 bits) v: 3.3 Mesa 18.3.6 compat-v: 3.1 
           direct render: Yes 
Audio:     Device-1: VIA VT8237A/VT8251 HDA driver: snd_hda_intel v: kernel 
           bus ID: 00:14.0 chip ID: 1106:3288 
           Sound Server: ALSA v: k4.9.176-antix.1-486-smp 
Network:   Device-1: Ralink RT2790 Wireless 802.11n 1T/2R PCIe driver: rt2800pci v: 2.3.0 
           port: f000 bus ID: 02:00.0 chip ID: 1814:0781 
           IF: wlan0 state: down mac: <filter> 
           Device-2: Realtek RTL-8110SC/8169SC Gigabit Ethernet driver: r8169 
           v: 2.3LK-NAPI port: bc00 bus ID: 04:06.0 chip ID: 10ec:8167 
           IF: eth0 state: down mac: <filter> 
Drives:    Local Storage: total: 5.61 GiB used: 2.27 GiB (40.5%) 
           ID-1: /dev/sda vendor: SanDisk model: SDCFH-004G size: 3.73 GiB 
           speed: <unknown> serial: <filter> rev: 6.02 scheme: MBR 
           ID-2: /dev/sdb type: USB vendor: Transcend model: JetFlash Transcend 2GB 
           size: 1.88 GiB serial: <filter> rev: 8.07 scheme: MBR 
Partition: ID-1: / size: 2.30 GiB used: 2.2 MiB (0.1%) fs: overlay source: ERR-102 
Sensors:   System Temperatures: cpu: 29.0 C mobo: N/A 
           Fan Speeds (RPM): N/A 
Info:      Processes: 128 Uptime: 1m Memory: 2.90 GiB used: 124.7 MiB (4.2%) 
           Init: SysVinit v: 2.93 runlevel: 5 default: 5 Compilers: gcc: 8.3.0 alt: 8 
           Shell: bash v: 5.0.3 running in: roxterm inxi: 3.0.33 

No special boot parameters were necessary.

Update: From second boot on the picture becomes pixelated with bad colours. Funny thing is I can only boot with the second DVI interface which gets this sort of corruption, but when having booted I can connect to the first DVI interface (closer to the CPU) and the picture is cristal clear. This box is dodgy and useless for graphical session, but it was not built for it.

• This reply was modified 1 year, 1 month ago by eugen-b.
• This reply was modified 1 year, 1 month ago by eugen-b.

Attachments:

1. Xorg.0.log_.openchrome.txt

[Forum Admin BitJam]

Xecure wrote:

With boot options, I still haven’t tested what “wicd” and “nowicd” options do.

They do nothing. We should have removed them. Were you booting via UEFI so the selection was in a text menu?

Will they now enable/disable Connman instead of WICD (no longer installed) so that ceni works properly, or are they to be changed for “connman” and “noconnman” in the future?

You can disable the connman service with the “disable=c” option (add “c” to the list of letters after “disable=”).

WIfi not working with Connman in first start

I had difficulty running connmanctl from the command line. I needed to make a little cheat sheet of the commands I need to run after I started connmanctl:

agent on
enable wifi
services
connect wifi_<tab>

Perhaps there are similar hoops you need to jump through in the GUI version so you don’t need to reboot. I’d love to add this small cheat sheet somewhere but I’m not sure where the best place is. I’m lost without it.[/quote]

Tsplash experience
No fails from tsplash for now. On laptop perfect. I really like it very much.
First time booting with the liveUSB on my tablet (which is slow but with an enormous screen resolution), in the step after loading to RAM, I think it displays “Start Init”, the screen went totally black. I waited for about 20 seconds and decided to see what was going on.
I hit ALT+F1 (as the tsplash screen recommended before it went black to see more info). The terminal was there (so all well) and the last command displayed said “waiting for /dev to fully populate”. I returned to the black screen with ALT+F10 and after a few seconds, the tsplash screen returned (I believe that with a change of resolution). So everything was good.

Alas, this is one weakness of the tsplash system. I can explain why it does this but I don’t have a quick or easy way to fix it. The root of the problem is that we try to keep the live system “lean and mean”. To that end we put as little extra stuff in the live initrd as possible because almost all of this already exists in the linuxfs file. IIRC the Debian initrd is 26M (or maybe larger) while our live initrd is only 4.3M. A major reason for this is we don’t include: udev, video drivers, video driver firmware, internet drivers, or internet driver firmware.

We rely on the udev service to load video (and other drivers) when it runs. By now most video drivers now are KMS (kernel mode setting or ‘modesetting”). This means the graphics driver also handles the virtual consoles. When these drivers load the console screen jumps around, can often get garbled, and the resolution often changes. This looks like s**t and was one of our main motivations to write tsplash.

We added a little code to the udev init.d script so when it starts, we first blank the screen. Then when it is done we unblank (redraw) the screen and say “udev done” or something like that. This does a great job at hiding all the jumping about and garble, but, as you know, the udev script can sometimes hang while waiting for /dev/ to populate so we end up with apparently hanging and a blank screen.

One way I tried to get around this (a bit) was to create a udev rule to redraw the tsplash screen when a new framebuffer is used. This would at least minimize the time the screen is blank and, if we are lucky, could give you a splash screen while we are waiting for udev if a KMS video driver gets loaded before udev hangs.

There were several problems with the udev rule so I finally dropped it but with some more work I might be able to add it back. Currently the tsplash screen is on tty10. In order to run a program to draw on tty10, I need to call it with the “openvt” command although I might be able to do the same thing with some IO redirection. But, for some reason, I’m not allowed to call openvt from a udev rule. Another problem is that the framebuffer gets set at least twice in rapid succession when the video driver loads (I don’t know why) so the usual case has three calls to redraw the screen in rapid succession, two from the udev rule and then the third from the end of the udev script (in case no KMS driver was loaded so our udev rule was never triggered).

So, for now, we are stuck with displaying a blank screen when udev hangs. I *might* be able to handle this better in the future but it might take some work and it might not make a difference. So for now, this falls into the “known bug” category.

Fantastic work, guys!

Thanks!

Context is worth 80 IQ points -- Alan Kay

[Member Xecure]

BitJam wrote:

…

So, for now, we are stuck with displaying a blank screen when udev hangs. I *might* be able to handle this better in the future but it might take some work and it might not make a difference. So for now, this falls into the “known bug” category.

Wow. Thanks for the detailed answer! I’ve learned a lot. It is amazing how much work goes into avoiding all those strange behaviors to achieve such great results. We really appreciate all your hard work! This new tsplash screen, and all the remodeled tools and boot-rescue options are fantastic. All this really makes antiX stand out from the other distros!

[Author]

Posts