Page 2 of 3

Re: Debian buster: are you testing?

Posted: 2019-01-05 15:46
by Zjho
Finally surrendered running Buster on spare box after BSODs with unresponsive cursor on all choices of kernel and rescue modes.
However looking forward to trying again when goes officially stable. Stretch is rock solid on main box (identical hardware).

Re: Debian buster: are you testing?

Posted: 2019-01-05 17:23
by Bulkley
Head_on_a_Stick wrote:@Bulkley, if you post the full output of

Code: Select all

apt-get -s dist-upgrade
then we may be able to help :)
Thanks for the offer. When I do a dist-upgrade I do it from a console (no xorg) as follows:

Code: Select all

apt-get update && apt-get upgrade
[change apt sources]
apt-get update && apt-get upgrade
apt-get dist upgrade
Is apt-get -s dist-upgrade useful without doing the first three steps?

As I'm back in Stretch I'll wait until the next urge. Actually I'm not upset about being stuck in Stretch. My system is running very well with everything set the way I like it. I could keep it like this for a long time.

My big attraction to Debian is and always has been package management. I clearly remember RPM dependency hell. :twisted: Lately I've been experimenting with a MX Linux 18 live-USB. As I understand it MX runs without Systemd controlling everything yet keeping Debian packages and package management. Now if I can find a way to dump Xfce and replace it with Openbox (not so easy on a live-USB) I'll take MX more seriously.

Re: Debian buster: are you testing?

Posted: 2019-01-05 20:04
by HuangLao

Re: Debian buster: are you testing?

Posted: 2019-01-05 20:25
by stevepusser
Bulkley wrote:
As I understand it MX runs without Systemd controlling everything yet keeping Debian packages and package management. Now if I can find a way to dump Xfce and replace it with Openbox (not so easy on a live-USB) I'll take MX more seriously.
Yes, it boots without systemd by default, though it is an option in the Advanced part of the GRUB menu. Just today, the other MX and antiX devs are fixing a weird problem with Debian's sysvinit scripts that prevent flatpaks like Steam, Atom, etc. form just working out of the box without systemd. Apparently, Debian just went the opposite way from the rest of the Linux world when setting a mount point symlink in /tmp/shm or something, and the flatpaks guys aren't accounting for that. It seems a pretty easy fix, though.

You might be able to get Openbox set up and remove xfce in a persistent MX Live USB, then use the snapshot tool to make your own Openbox respin ISO. I know the snapshot tool can do that, but aren't that familiar with the persistent Live option, just that it's the most advanced there is. Or you could make a respin in a virtual machine, too. Asking on the MX forums would probably help a lot.

Re: Debian buster: are you testing?

Posted: 2019-01-05 21:39
by Bulkley
HuangLao, Thanks for the links.

My musings are side tracking the thread. I think I should start a fresh one.

Re: Debian buster: are you testing?

Posted: 2019-01-05 23:08
by Head_on_a_Stick
Bulkley wrote:When I do a dist-upgrade I do it from a console (no xorg) as follows:

Code: Select all

apt-get update && apt-get upgrade
[change apt sources]
apt-get update && apt-get upgrade
apt-get dist upgrade
Is apt-get -s dist-upgrade useful without doing the first three steps?
Yes, it should still return useful information :)

The trick is to point the sources to buster, run `apt update` then `apt-get -s dist-upgrade` then revert the sources back to stretch and `apt update` again to bring your box back to stable.

Re: Debian buster: are you testing?

Posted: 2019-01-06 01:58
by Bulkley
apt-get -s dist-upgrade produced 126 pages (in LibreOffice Writer) of stuff and is much to big to post here. Just the conf part occupies 40 pages. I browsed through it and did not see any error messages. It all looks routine.

When I tried the dist-upgrade the other day I saw no error messages until I rebooted.

Re: Debian buster: are you testing?

Posted: 2019-01-06 10:30
by Head_on_a_Stick
Bulkley wrote:When I tried the dist-upgrade the other day I saw no error messages until I rebooted.
Ah, I see.

I'm happy to help you troubleshoot that if you want, what did the error messages say?

Re: Debian buster: are you testing?

Posted: 2019-01-07 21:13
by Bulkley
Head_on_a_Stick wrote:I'm happy to help you troubleshoot that if you want, what did the error messages say?
Thanks for the offer. I think I know where to start: hardware. This machine is at least ten years old. At that, I got it second hand and changed a few things to suit my needs. The mobo is original. Buster is most likely using a kernel and/or other software that doesn't look too far backward. I made up a live Buster USB and it would not boot. I tried two different versions and nothing. It's simply not recognized. It looks like that eufi (sorry I can't remember acronyms) setting in the BIOS. I'll look into it when I have time. I'm in no hurry to do that. As I have mentioned elsewhere, I like my Stretch as is.

Re: Debian buster: are you testing?

Posted: 2019-01-08 04:32
by xepan
Bulkley wrote: Thanks for the offer. I think I know where to start: hardware. This machine is at least ten years old. At that, I got it second hand and changed a few things to suit my needs. The mobo is original. Buster is most likely using a kernel and/or other software that doesn't look too far backward.
None of my PC's or laptops is younger than 10 years, besides the raspberries, and actual kernels and distributions work well on them.
And i got quite some, 6 or such (and quite a few i threw away the last years, so i can speak for more than just those 6).

Might still be that in your case that is the problem, i wouldn't know. I am not much of a reader, more of the "let's try and see" guys.

Code: Select all

# dmidecode 3.2
Getting SMBIOS data from sysfs.
SMBIOS 2.4 present.

Handle 0x0002, DMI type 1, 27 bytes
System Information
        Manufacturer: Hewlett-Packard
        Product Name: HP Compaq dc5750 Small Form Factor
        Version:
        Serial Number: CZC71856VC
        UUID: d706ceac-48e2-db11-bbda-4b4896d9001a
        Wake-up Type: Power Switch
        SKU Number: EW318AV
        Family: 103C_53307F

# grep "model name" /proc/cpuinfo
model name      : AMD Athlon(tm) 64 X2 Dual Core Processor 4200+
model name      : AMD Athlon(tm) 64 X2 Dual Core Processor 4200+
# uname -a
Linux voidlinux 4.19.13_1 #1 SMP PREEMPT Sun Dec 30 08:54:53 UTC 2018 i686 GNU/Linux
#
one of my two machines which are able to handle 64 :-) ...

Re: Debian buster: are you testing?

Posted: 2019-01-08 05:46
by llivv
xepan wrote: None of my PC's or laptops is younger than 10 years, and actual kernels and distributions work well on them.
Same here...
I have 15 year olds I don't use much anymore, and wouldn't you know it, when I went to use one of them for a hybrid car issue, I found a leaking cap next to the PCIx slot and the onboard video failed. 'rethink the hybrid car maintenance' ....
So I got a used 2008 mobo/processor combo [$25] and dropped it in the case after pulling the toasted mobo out. And that is what I'm using now. Intel DG41RQ. Xeon X3300, 2GB ddr2 pc6400, 500 GB Western Digital SATA 300 .
I upgraded the processor from a core2 duo 2.4 Ghz to the core2 quad 2.6Ghz [another $25]
and upgraded the 512mB ddr2 pc5300 to 2Gb ddr2 pc6400 [$15] / available computer funds.
It's old trash compared to some of the $5 or 6K machines asking for help in this forum.

Re: Debian buster: are you testing?

Posted: 2019-01-08 05:49
by xepan
Well, it would probably be an idea to upgrade one or two of them (more i don't use regulrarly anyway).
Thing is: they do what i want, so i don't (i am always broken).
modern webbrowsers don't really work, but besides text-only websites like forums i don't use the web. into the bargain: no need to block ads :-)

Just chat, not arguing that upgrading sure makes sense.

Re: Debian buster: are you testing?

Posted: 2019-01-08 06:03
by llivv
I just don't get why so many need to [upgrade to newest models] as soon as they are released.
Unless they want to support the R&D for quantum automation[cars, phones, computers, TV's] are the big ones, but almost everything nowadays.

I sure don't want to support that, such a waste.

Re: Debian buster: are you testing?

Posted: 2019-01-08 14:32
by None1975
Head_on_a_Stick wrote:The trick is to point the sources to buster, run `apt update` then `apt-get -s dist-upgrade` then revert the sources back to stretch and `apt update` again to bring your box back to stable.
You can go from stable --> testing --> unstable. But the reverse direction is not "possible". Actually, if you are an expert and if you are willing to spend some time and if you are real careful and if you know what you are doing, then it might be possible to go from unstable to testing and then to stable. The installer scripts are not designed to do that. So in the process, your configuration files might be lost and...

Re: Debian buster: are you testing?

Posted: 2019-01-08 15:35
by Bulkley
None1975 wrote:
Head_on_a_Stick wrote:The trick is to point the sources to buster, run `apt update` then `apt-get -s dist-upgrade` then revert the sources back to stretch and `apt update` again to bring your box back to stable.
You can go from stable --> testing --> unstable. But the reverse direction is not "possible". . . .
Note the '-s' which is just a look-see.

Re: Debian buster: are you testing?

Posted: 2019-01-08 18:04
by Head_on_a_Stick
None1975 wrote:if you are an expert and if you are willing to spend some time and if you are real careful and if you know what you are doing, then it might be possible to go from unstable to testing and then to stable
Guide here: https://www.debian.org/doc/manuals/debi ... owngrading :mrgreen:

But yes, as Bulkley notes, we were using the --simulate flag to get a handle on the errors without b0rking the box.

Re: Debian buster: are you testing?

Posted: 2019-01-08 19:10
by None1975
Head_on_a_Stick wrote:But yes, as Bulkley notes, we were using the --simulate flag to get a handle on the errors without b0rking the box.
Now I understood. Thanks for the explanation.

Re: Debian buster: are you testing?

Posted: 2019-01-08 22:06
by Bulkley
xepan wrote:None of my PC's or laptops is younger than 10 years, besides the raspberries, and actual kernels and distributions work well on them.
This one has an odd BIOS that doesn't like live-USBs. For each one I have to manually set BIOS to the front of the boot cycle or it won't find it. It will run them begrudgingly but they have to be perfect. Anyway, I tried both of the Buster live-USBs on another machine and both reported not bootable. Curiously, I made a Devuan live-USB with the same tools and it booted easily. I have made live-Debian USBs for Stretch, Jessie, etc. The computer demons just don't want me to test Buster. :lol:

Re: Debian buster: are you testing?

Posted: 2019-01-09 05:15
by xepan
Ah, got you.
Some of mine don't even boot from USB :-)

I first thought you thought about throwing it away (and my point was: perhaps ! it is still usable. You last comment seems to confirm that, but simply not with buster - right now).
You don't seem to have wanted that, throwing it away, at all, so ... well: my fault.

-
been a while, but: In the few cases where debian didn't boot, antiX often was a good shot for me. On some i couldn't even make that run.
but as said: been a while, and many things have changed. I got no CD's ...

Re: Debian buster: are you testing?

Posted: 2019-01-09 06:03
by Bulkley
xepan wrote:In the few cases where debian didn't boot, antiX often was a good shot for me.
Yup. Antix boots. I spent a couple hours today on a MX live-USB trying to figure out the mechanics of how it differs from Debian. The Antix/MX crew is quite innovative. I tried FreeBSD but never got to a gui. You win some and lose some.