SlackwareThis Forum is for the discussion of Slackware Linux.
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
Dec 25 21:31:29 UTC 2020 <snip> l/pulseaudio-14.0-x86_64-3.txz: Rebuilt.
Patched to support detecting elogind at compile time when built with meson.
This restores PulseAudio exiting upon user logout.
Thanks to LuckyCyborg for the bug report and patch.
Thanks LuckyCyborg and Pat. Incidentally I prefer not to start PulseAudio with the session, only when an app requests it, but that's somehow off topic.
Dear Mr. Spaier, honestly I would love to exists an elogind feature which permits it to handle "user target" daemons like is PulseAudio, then to start them on an user login and to stop them on the user logout.
Why? That will heavily simplify my struggle with the PipeWire daemons, which has yet? no way to be aware that an user login session ended, and it's time for them to exit.
Unfortunately, there is no such thing yet, and my patch proposal essentially just restores the PulseAudio behavior as was in a CK2 driven system - quickly exiting on user logout.
And for the same unfortunate reason we're about to invent an "elogind aware daemon supervisor" ... Hopefully.
Anyway, I hope that I can give you an useful hint: the PipeWire has no auto-spawn features, and it is supposed to be launched when user logins. And to stay alive until user logouts.
In my humble opinion, maybe is a good idea to consider carefully this particular behavior of PipeWire and to avoid now that auto-spawning at all costs, as PipeWire is the PulseAudio successor - in few months it even will already replace PulseAudio server in Fedora, and trust me: it will NOT be optional.
Last edited by LuckyCyborg; 12-26-2020 at 07:52 AM.
Why not kill pulseaudio in .bash_logout?
/bin/killall pulseaudio 2> /dev/null
This way cannot be used, because of some caveats:
1. The "user target" services like is PulseAudio should be stopped when the last login session closes. Simply to image, considering that the user can login in three consoles, from where the logouts may happen in whatever order - then the kill should happen when the last one is closed.
True, you may ask elogind - and so does even PulseAudio to find out this information, with no needs for supplementary scripting.
2. There is no bash login (and consequently no final bash logout) when the runlevel 4 is used - together with SDDM or whatever other DM.
BTW, the PulseAudo had this feature of "exiting on user logout" since long time while was used ConsoleKit2 - even into Slackware 14.2 is this particular PulseAudio feature.
And I seen no one protesting against it.
So, why we should NOT have it back now, when we use elogind, instead of ConsoleKit2?
Last edited by LuckyCyborg; 12-26-2020 at 05:47 PM.
1. The "user target" services like is PulseAudio should be stopped when the last login session closes. Simply to image, considering that the user can login in three consoles, from where the logouts may happen in whatever order - then the kill should happen when the last one is closed.
True, you may ask elogind - and so does even PulseAudio to find out this information, with no needs for supplementary scripting.
2. There is no bash login (and consequently no final bash logout) when the runlevel 4 is used - together with SDDM or whatever other DM.
BTW, the PulseAudo had this feature of "exiting on user logout" since long time while was used ConsoleKit2 - even into Slackware 14.2 is this particular PulseAudio feature.
And I seen no one protesting against it.
So, why we should NOT have it back now, when we use elogind, instead of ConsoleKit2?
I for one consider a tidy state of a running system quite high on the priority list - having various "services" be run and shut down orderly at all times and in all cases is something Slackware is renowned since it's inception - the "slack" in its name implies exactly that.
Also, recent GNU/Linux-es resorting to less transparent systems like message buses and the accompanying frameworks supporting that run quite the opposite way - making simple tasks awkward and tedious to master and comprehend, so there it is i said it out loud.
So, yes, if we opted for elogind, let's make the most use of it for all things practical ad let's master it the best we can?
[ 13.079598] atomisp: module is from the staging directory, the quality is unknown, you have been warned.
[ 13.118568] **********************************************************
[ 13.120732] ** NOTICE NOTICE NOTICE NOTICE NOTICE NOTICE NOTICE **
[ 13.122932] ** **
[ 13.125064] ** trace_printk() being used. Allocating extra memory. **
[ 13.127197] ** **
[ 13.129337] ** This means that this is a DEBUG kernel and it is **
[ 13.131492] ** unsafe for production use. **
[ 13.133647] ** **
[ 13.135883] ** If you see this message and you are not debugging **
[ 13.138098] ** the kernel, report this immediately to your vendor! **
[ 13.140391] ** **
[ 13.142672] ** NOTICE NOTICE NOTICE NOTICE NOTICE NOTICE NOTICE **
[ 13.144963] **********************************************************
[ 13.150004] atomisp-isp2 0000:00:03.0: Support for Cherrytrail (ISP2401) was disabled at compile time
As instructed by the fresh upgraded Kernel 5.10.x which unconditionally demands me to contact immediately the vendor:
Minnesota, we have a problem!
Last edited by ZhaoLin1457; 12-27-2020 at 05:25 AM.
[ 13.079598] atomisp: module is from the staging directory, the quality is unknown, you have been warned.
[ 13.118568] **********************************************************
[ 13.120732] ** NOTICE NOTICE NOTICE NOTICE NOTICE NOTICE NOTICE **
[ 13.122932] ** **
[ 13.125064] ** trace_printk() being used. Allocating extra memory. **
[ 13.127197] ** **
[ 13.129337] ** This means that this is a DEBUG kernel and it is **
[ 13.131492] ** unsafe for production use. **
[ 13.133647] ** **
[ 13.135883] ** If you see this message and you are not debugging **
[ 13.138098] ** the kernel, report this immediately to your vendor! **
[ 13.140391] ** **
[ 13.142672] ** NOTICE NOTICE NOTICE NOTICE NOTICE NOTICE NOTICE **
[ 13.144963] **********************************************************
[ 13.150004] atomisp-isp2 0000:00:03.0: Support for Cherrytrail (ISP2401) was disabled at compile time
As instructed by the fresh upgraded Kernel 5.10.x which unconditionally demands me to contact immediately the vendor:
Minnesota, we have a problem!
I think here is a related discussion (the whole thread, actually)
The udev rules file from sane needs a higher priority. It sets ENV{libsane_matched}="yes" and is currently named 80-libsane.rules – that's too late for 70-uaccess.rules and some others (69-libmtp.rules, 56-hpmud.rules and maybe more) which check this variable to e.g. add ACLs for the currently logged-in user. The original libsane.rules is installed as 49-sane.rules by Gentoo Arch and as 41-libsane.rules by Arch Gentoo.
[Edit: Mixed up Gentoo/Arch filenames, thx @gmgf!]
Additionally there is still a little issue with mariadb since we have PAM: /etc/security/user_map.conf is not installed as a .new file so user-changes to this file will be overwritten on update!
Last edited by Markus Wiesner; 12-27-2020 at 12:13 PM.
Actually in 'Arch' current (with lasted packages), the file '80-libsane.rules' it's, named, '49-sane.rules', but in openSUSE, it's named like, slackware.
It's probably time to reassess upgrading fluidsynth to 2.x. That's been the version up on SlackBuilds.org for a few months now, and it seems like everything works with minimal intervention. I'm not sure what all depends on it in -current, or I'd play around with it in a VM for you.
Since -current already enabling zstd in the kernel for compressing initrd, why not using it as default in mkinitrd and also remove some of compression type:
I've been using zstd as initrd compression (replacing gzip -9 in /sbin/mkinitrd with ztsd -19) for last two month. Initrd size is smaller than gzip-based and boot time is significantly reduced with the same hardware.
About initrd
I've been using zstd as initrd compression (replacing gzip -9 in /sbin/mkinitrd with ztsd -19) for last two month. Initrd size is smaller than gzip-based and boot time is significantly reduced with the same hardware.
It's just need a small patching of grub /etc/grub.d/10_linux adding "initrd.xz" "initrd-${version}.xz" to add the option in grub-mkconfig.
EDIT :
Two patch files added
Last edited by Tonus; 12-28-2020 at 10:31 AM.
Reason: Add patchs to 10_linux and mkinitrd
This is a new major release of TigerVNC, but also a security release. Users that rely on the TLS feature in the viewers are recommended to upgrade as soon as possible.
fltk-1.3.5 (fltk required by tigervnc) is still release - no update needed.
Requested previously, re-requesting in case it was missed.
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.
