Posts: 18
greywolf
Joined: 02 Feb 2014
#16
i fear the hand of systemd here, but I can't prove it.

edit: btw, this was testing with a wired interface, so I don't think wireless is really the culprit here anymore.
Exactly!!! I similarly suspect the little blighter has snuck in during a dist-upgrade somehow; but I have not had time to track it down.

I did an install that ignored the choice of 'testing' and installed 'Jessie/stable'. That's ok for now I thought. I did my 'basic' setup routine that I do post almost any install; then, when I had stuff looking the way I wanted & my essentials installed, I installed a Liquorix kernel (4.04) and did a dist-upgrade to 'testing' from outside X. Changed antix & debian repos only. After reboot, I had no sound AT ALL. Standard Intel onboard sound; never gives a problem. I looked very quickly at the upgrade log and thought I tried all the normal checks mentioned here and elsewhere for sound trouble. Nothing changed it.

After deciding I was going to reinstall and leave it on Jessie, I thought, why not try pulseaudio before I wipe it. Of course that brought in a swag of systemd stuff so I can no longer prove whether it was there before or not. However, sound worked without a problem, so the dist-upgrade somehow killed alsa?!

As a separate matter, bringing in a heap of systemd seemed to screw with the graphics under certain themes.

I am no longer optimistic that running without systemd installed is feasible at any level above Jessie? Just too many variables & inter-dependent changes to catch you up?

Anyway, FWIW, antix (spacefm/fluxbox) using stable is working just fine at the moment.

greywolf.
Posts: 2,238
dolphin_oracle
Joined: 16 Dec 2007
#17
greywolf wrote:
i fear the hand of systemd here, but I can't prove it.

edit: btw, this was testing with a wired interface, so I don't think wireless is really the culprit here anymore.
Exactly!!! I similarly suspect the little blighter has snuck in during a dist-upgrade somehow; but I have not had time to track it down.

I did an install that ignored the choice of 'testing' and installed 'Jessie/stable'. That's ok for now I thought. I did my 'basic' setup routine that I do post almost any install; then, when I had stuff looking the way I wanted & my essentials installed, I installed a Liquorix kernel (4.04) and did a dist-upgrade to 'testing' from outside X. Changed antix & debian repos only. After reboot, I had no sound AT ALL. Standard Intel onboard sound; never gives a problem. I looked very quickly at the upgrade log and thought I tried all the normal checks mentioned here and elsewhere for sound trouble. Nothing changed it.

After deciding I was going to reinstall and leave it on Jessie, I thought, why not try pulseaudio before I wipe it. Of course that brought in a swag of systemd stuff so I can no longer prove whether it was there before or not. However, sound worked without a problem, so the dist-upgrade somehow killed alsa?!

As a separate matter, bringing in a heap of systemd seemed to screw with the graphics under certain themes.

I am no longer optimistic that running without systemd installed is feasible at any level above Jessie? Just too many variables & inter-dependent changes to catch you up?

Anyway, FWIW, antix (spacefm/fluxbox) using stable is working just fine at the moment.

greywolf.
I think part of the problem in testing is that packages expect systemd to be available, and antiX ships with the apt settings such that systemd will not come down.

Things might settle down in a while as testing/stretch matures.
Posts: 325
male
Joined: 04 Nov 2011
#18
What you write here, I can not understand!

antiX has chosen V15 against systemd.

Everyone knows here, Debian has decided with Jessie for systemd.

The activation of the testing repos spĆ¼hlt course systemd packages in the system.

Have they thought systemd cares about adjusting the sysvinit-scripts in antiX?
Sooner or later they will experience the same with stable...
Posts: 1,139
masinick
Joined: 26 Apr 2008
#19
dolphin_oracle wrote:removing my eth0 interface from / etc/network/interfaces and then just allowing the wicd-daemon to handle the configuration survived reboot. The network interface worked as expected.
Nice diagnosis on this matter. I have not yet come across this issue, but I have multiple antiX and MX instances on two different portable systems; some of them are running off the Stable repos and may not be exercising this particular code path, but I have at least one instance somewhere that is or was using Testing, and at one point, I had one or two instances that were using Sid. So eventually I may come across a similar issue and if I do, this is really good information, so I certainly appreciate it!