Search Results for 'ppa'

[Member olsztyn]

stevesr0 wrote:

ot sure if this is still an issue, but there was a problem last year (17 January 2022) on the Artix forum with Bluetoothd causing Wireplumber not to work after a suspend. Apparently disabling bluetoothd as a runit service or using the following code fixed it:

Thanks @stevesr0…
It does not appear that issue with wireplumber in my repeated tests is related to bluetooth and suspend reason. It rather seems to be related to pipewire components startup timing:
– If pipewire components are started as in the specified above sequence with no timing (sleep) between them, then wireplumber fails to start in 30% cases (sample is small – just 10 reboots, so 30% might be incorrect value).
– If pipewire components are started as above but with ‘sleep 5’ timing space between them then wireplumber starts every time. Again my disclaimer due to small sample of 10 reboots.

So assuming that injecting such timing delays between starting pipewire components makes wireplumber/pipewire start and work every time, this might be considered the way to initalize pipewire infrastructure.

I must mention again that such discovery of timing delays (sleep 5) between start of each pipewire component is not mine but credit for this goes to @Caprea. I just followed this idea.

Pipewire sound and bluetooth support seams to work fine if all components, including wireplumber are initialized. The entire memory footprint toll in my setting (Thinkpad X220, i5, 4Gb) appears to me just about 20Mb over Alsa base, so not that much and probably even less than Pulseaudio, the way I remember. SO it appears to be a viable sound system to use, if all the claims about lower latency are indeed correct.

Sound quality – I do not hear any difference, but I am not the right person to notice. I just remember a post about sound quality test performed by @calciumsodium, where two people found Pulseaudio sound quality to be slightly better.

As the necessity to inject timing delays (about sleep 5) between startup of pipewire components are necessary for wireplumber, this is a workaround to me, rather than a solid and managed sequencing of these components, so I think this might confirm why @Xecure was using Runit user services to start pipewire pieces instead.
However not having details I am not able to do it so far and I have a feeling it is not simple either.
Looks to me that it should be the responsibility of wireplumber design to ensure the preceding required components are fully up but apparently it is not the case…

• This reply was modified 2 months, 1 week ago by dolphin_oracle.
• This reply was modified 2 months, 1 week ago by dolphin_oracle.
• This reply was modified 2 months, 1 week ago by dolphin_oracle.

Live antiX Boot Options (Previously posted by Xecure):
https://antixlinuxfan.miraheze.org/wiki/Table_of_antiX_Boot_Parameters

[Moderator Brian Masinick]

stevesr0 wrote:

Hi Olsztyn,

Not sure if this is still an issue, but there was a problem last year (17 January 2022) on the Artix forum with Bluetoothd causing Wireplumber not to work after a suspend. Apparently disabling bluetoothd as a runit service or using the following code fixed it:

Code
# ~/.confi/Wireplumber/Wireplumber.conf
…
  { name = policy.lua, type = config/lua }
# { name = Bluetooth.lua, type = config/lua }
]

stevesr0

Hmm, that is an interesting issue, because when I’m using sound, though in some cases I can plug in my headphones, they are designed to work with Bluetooth, and in fact, for the past couple of years when I HAVE had an issue with a Debian-based distribution, it’s usually been with Bluetooth connectivity. I have had some success; however when I do have Bluetooth issues, at least 90% of them have been with Debian-based distributions, and antiX has suffered more than many others because of upstream software issues with the stack. If there is any reason that I’d start using pipewire, wireplumber, etc., it would probably be to get a “win” with this stuff more often, so I hope these matters improve greatly as the Debian Bookworm and the antiX 23 projects continue development.

I may check around Debian from time to time to see what’s happening, and if there is an opportunity to TEST upstream, I may do so, since I use 3-4 Debian-based distributions, including Debian itself, and of COURSE, antiX!

--
Brian Masinick

[Member stevesr0]

Hi Olsztyn,

Not sure if this is still an issue, but there was a problem last year (17 January 2022) on the Artix forum with Bluetoothd causing Wireplumber not to work after a suspend. Apparently disabling bluetoothd as a runit service or using the following code fixed it:

Code
# ~/.confi/Wireplumber/Wireplumber.conf
…
  { name = policy.lua, type = config/lua }
# { name = Bluetooth.lua, type = config/lua }
]

stevesr0

[Member 1040STE]

I encounter difficulties in mounting my NAS with Sambashare.
The problem is : Antix tries to establish an SMB1 connection, and the Synology denies this.
No, I don’t want to allow SMB1 connections 🙂
How can I force SMB2 or 3 in Antix ? I find here and there solutions implying modifying smb.conf, but I understand Samba isn’t “really” installed on Antix (no smb.conf nor related folder).
And apparently installing samba fails because some dependencies can’t be installed.
Thanks for your help!

[Moderator Brian Masinick]

Eek!

Too many experiments apparently.
Glad you spotted it, thanks!

--
Brian Masinick

[Moderator Brian Masinick]

olsztyn wrote:

Just to notice, the number of posts in this thread exceeded 1,000… The longest thread ever.
Remarkable popularity…

Once upon a time it was only me.
Apparently some people other than me also like to share information on their own systems. Not sure if anyone else is tracking change here or not.

A long time ago on another long lost forum we had a thread, something like —
“What are you here with and Whatever?” and it was a wide open topic when a discussion didn’t fit anywhere else.

I don’t want that here, but I don’t mind sharing this topic as long as it’s primarily about the information gathering about our systems.

--
Brian Masinick

[Moderator Brian Masinick]

I thought I had taken care of that; apparently I did not.
Works now thanks!

Generating grub configuration file ...
Found linux image: /boot/vmlinuz-6.1.10-antix.1-amd64-smp
Found initrd image: /boot/initrd.img-6.1.10-antix.1-amd64-smp
Found linux image: /boot/vmlinuz-4.19.0-256-antix.1-amd64-smp
Found initrd image: /boot/initrd.img-4.19.0-256-antix.1-amd64-smp
Found mtest-64.efi image: /boot/uefi-mt/mtest-64.efi

--
Brian Masinick

[Moderator Brian Masinick]

rej wrote:

Hi Brian-

Thanks for your help.

When creating the snapshot it has the option to change the name of the file, and I just recently started using that feature, not realizing, that when the name is changed, it probably no longer recognizes the snapshot as “MX” or “antiX” and goes to “dd” automatically.

I had not connected the two until today.

Thank you for taking the time to do the tests and provide feedback – I appreciate it!

So did that effectively solve your issue? What I’ve consistently seen is that I can always create either antiX or MX Linux images using the Live USB Maker tool and it recognizes images from these two distributions, but given your discovery about the names, I suspect there is something in the code that examines whether or not a Full-featured mode is selected – and apparently the choice is based on the “distribution-provided” identifiers; I did not know that, but the other reason is that the built-in, recognized capabilities are in those two distributions. While there may be other distributions capable of certain features beyond what dd offers, there’s no way to ensure that, hence the reason why other named distributions are copied with dd, which is simply a byte for byte copying tool.

--
Brian Masinick

[Moderator Brian Masinick]

From https://kernel.org/category/releases.html

Version 	Maintainer 	                Released 	Projected EOL
6.1 	Greg Kroah-Hartman & Sasha Levin 	2022-12-11 	Dec, 2026
5.15 	Greg Kroah-Hartman & Sasha Levin 	2021-10-31 	Oct, 2026
5.10 	Greg Kroah-Hartman & Sasha Levin 	2020-12-13 	Dec, 2026
5.4 	Greg Kroah-Hartman & Sasha Levin 	2019-11-24 	Dec, 2025
4.19 	Greg Kroah-Hartman & Sasha Levin 	2018-10-22 	Dec, 2024
4.14 	Greg Kroah-Hartman & Sasha Levin 	2017-11-12 	Jan, 2024

As I’ve said before, some people do well with the 5.10 kernel and it was known to fix a few difficult issues.
On the other hand, I know personally that the 5.10 kernel was also one of the most difficult kernels for
other systems, so while it apparently “fixed” some, it broke others that had otherwise worked for many years.

Here’s a convenient location to grab kernels for anyone wanting to either build source or directly use
a tarball:
https://www.kernel.org/feeds/kdist.xml

To me, the tree that needs OLD kernel support the most is the 32 bit tree.

For the 32 bit version – that’s the onethat we’ll have to work hard to keep running,
I think the 4.19 kernel would be a definite keeper, but I’m hoping some hacker
somewhere does some work beyond what kernel.org does and somehow keeps older,
unsupported kernels working. If yes, maybe that’s a potential solution.
Otherwise, when 4.19 support ceases, either the 5.4 or 5.10 kernel will have
to make due for those people. I’m concerned about what “Bookworm” changes will do to
impact that community, and for them, I think one of our legacy releases with lifespan
remaining may be their last place for “safe” use; after that, use those systems only
for simple work and avoid anything that needs to remain “secure”.

--
Brian Masinick

[Moderator Brian Masinick]

From https://kernel.org/category/releases.html

Version 	Maintainer 	                 Released 	Projected EOL
6.1 	Greg Kroah-Hartman & Sasha Levin 	2022-12-11 	Dec, 2026
5.15 	Greg Kroah-Hartman & Sasha Levin 	2021-10-31 	Oct, 2026
5.10 	Greg Kroah-Hartman & Sasha Levin 	2020-12-13 	Dec, 2026
5.4 	Greg Kroah-Hartman & Sasha Levin 	2019-11-24 	Dec, 2025
4.19 	Greg Kroah-Hartman & Sasha Levin 	2018-10-22 	Dec, 2024
4.14 	Greg Kroah-Hartman & Sasha Levin 	2017-11-12 	Jan, 2024

As I’ve said before, some people do well with the 5.10 kernel and it was known to fix a few difficult issues.
On the other hand, I know personally that the 5.10 kernel was also one of the most difficult kernels for
other systems, so while it apparently “fixed” some, it broke others that had otherwise worked for many years.

Here’s a convenient location to grab kernels for anyone wanting to either build source or directly use
a tarball:
https://www.kernel.org/feeds/kdist.xml

To me, the tree that needs OLD kernel support the most is the 32 bit tree.
I’m not 100% positive of this, but based on my personal and recent experience with systems between 7
and about 12-14 years old the Version 6 kernels HAVE been working on most, if not all, of the systems
I have in hand. The newest of these, my Dell Inspiron 5558, already quite a few years old, runs
VERY WELL with the 6.1 kernel.

Too bad the 4.9 kernel has ended support, it ran REALLY well on several of my older systems, though
the 4.19 kernel is also working well.

Someone made a suggestion about the 6.1 kernel – in the next, or one of the test versions before
release, let’s try having a 6.1 kernel as the default and see how well it works out, and hang
onto the 4.19 kernel for those it doesn’t work out.

Then for the 32 bit version – which I’ll also mention in the 32 bit thread, that’s the one
that we’ll have to work hard to keep running, I think the 4.19 kernel would be a definite
keeper, but I’m hoping some hacker somewhere does some work beyond what kernel.org does
and somehow keeps older, unsupported kernels working. If yes, maybe that’s a potential
solution. Otherwise, when 4.19 support ceases, either the 5.4 or 5.10 kernel will have
to make due for those people. I’m concerned about what “Bookworm” changes will do to
impact that community, and for them, I think one of our legacy releases with lifespan
remaining may be their last place for “safe” use; after that, use those systems only
for simple work and avoid anything that needs to remain “secure”.

--
Brian Masinick

[Forum Admin anticapitalista]

slim/slimski run X as root.
startx should run X as user.
Apparently,gdm or whatever it is called now, starts X as user,

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

antiX with runit - leaner and meaner.

[Moderator Brian Masinick]

stevesr0 wrote:

Just now (Eastern U.S. time 1701 on Sunday 5 Feb), several brief 503s.

stevesr0

I’m still suspicious of network storms, sudden mass intrusions specifically intended to halt and interrupt forum activities. I don’t know why but we’re targeted significantly more than our MX Linux friends because I have participated in a conversation with their moderator(s) and our webmaster and we are apparently a target. Why, we have no idea.

The changes we’ve made have improved restart times and they’re automatically performed when the site loses service. Hopefully it takes not much more than two minutes to recover.

--
Brian Masinick

[Moderator Brian Masinick]

Responsiveness and appearance are very positive, especially considering that I am running live.
In my previous run I specified toram; not sure I did so this time, but I’ve run non UEFI with the 4.9 kernel and UEFI with the 5.10 kernel and both of them are working OK, even though I’ve had problems on my Dell with this same 5.10.142 image; when I run from the USB I have no problem, so apparently the issues I was having were introduced through several kernels in use plus several distributions; here on the Live Alpha, it’s actually doing pretty well except for the package installer.

I even got around a little of that by adding in the MX repo and getting Firefox 109 from there!
I wonder if I can “steal” anything else during my test! Seriously, this is a pretty good Alpha; I’ve bragged about our early builds in the past; this one boots, looks good, isn’t perfect, yet it’s functional and has great looks!

We’re off to a good start.

--
Brian Masinick

[Forum Admin BitJam]

kaye wrote:

1. About creating live USB – most if not all of my live USB are of MBR partitioning scheme (Gparted calls it msdos I think). But apparently I can make them GPT as well, because antiX’s Live USB Maker provides the option to do so. Am I correct?

Correct.

2. If answer to number 1 is Yes, why would one choose to create an MBR-partitioned live USB of any linux distro, instead of GPT? Is it specifically for when the motherboard is old and is not of the UEFI type (which means the motherboard still uses BIOS?), and therefore cannot boot a GPT-partitioned live USB?

Short answer: GPT partitioning for a live-usb fails to work on several models of Dell machines.

Long answer: Live-usb-maker was built in conjunction with Ramsey from OSDisc. It started as a simple script to let Ramsey create “ideal” antiX and MX live-usbs. He had massive amounts of experience with live booting Linux since the business of OSDisc was to sell live booting CDs, DVDs, and usbs. Ramsey discovered that for a small subset of Dell machines, if you used GPT partitioning for a live-usb then you could either boot legacy or boot UEFI but not both. Whether legacy worked or UEFI worked depended on some partition flag or something like that.

So we use crappy msdos partitioning as the default in order to boot on as many systems as possible.

3. If the live USB is MBR-partitioned, I can use it to install the operating system onto a hard drive that is GPT-partitioned, correct?

Correct. All four combinations of mosdos and GPT partitioning work. It’s analogous to filesystems. The filesystem on the installed system is independent of the filesystem on the live system.

4. If the live USB is GPT-partitioned, I can use it to install the operating system onto a hard drive that is MBR-partitioned, as long as the motherboard uses UEFI and therefore can boot a GPT-partitioned live USB, correct?

UEFI booting and GPT partitioning are two different independent things. All four combinations of (UEFI, legacy) booting and (GPT, msdos) partitioning work. Our live-usb is proof of this fact. Our msdos partitioned live-usb boots both legacy and UEFI. So does our GPT partitioned live-usb.

The one caveat is that in order to install a UEFI bootloader you need to boot the live system in UEFI mode.

Again, think of the partitioning (msdos or GPT) like a filesystem. It’s independent of everything else. You do not need a UEFI capable machine to use GPT partitioning.

Context is worth 80 IQ points -- Alan Kay

[Member HisWord1st]

Brian Masinick wrote:

In whatever application you are using, can you select the text, starting at the beginning to wherever you want, then hit the right mouse button.

Does a list of choices appear?
If yes, select cut, copy or whatever you need.

Then navigate to whatever you want and right click again and paste the contents.

There are Ctrl sequences too but apparently they don’t work for you so this is another way to get it done.

Well, I’m in file manager SpaceFM trying to move images. The left-click drag and drop doesn’t work and when I right click the option to copy the image isn’t available, only to copy which copies image text, presumably filename.

I’m just using zzzFM now which is working for drag and drop but it would be great if I could set my PC to have these preferred settings rather than having it go to default every reboot…is there a way to do that?