Jarkko Sakkinen
Posts
4941Following
327Followers
492OpenPGP: 3AB05486C7752FE1
@gnubyte @pid_eins It's actually pretty good. I've tried it for my test VM's running archlinux while testing some kernel patch sets. My Fedora 40 host installation does not have it yet (at version 255).
In addition, it's a pretty nice stick for knocking the ice when testing various configurations related to polkit and PAM.
I've only recently moved my kernel QA from dusty old busybox to systemd (because I noticed that in some kernel features I was lagging over two years of user space progress), and it is definitely one of the more useful features of systemd for my work.
You can also pretty much use anything you want in the foreseeable future. In the end of the day, all of these are roughly front-ends configuring privileged access based on PAM, PolKit, MAC and DAC in varying levels of detail. There's also some other similar tools such as "really".
In addition, it's a pretty nice stick for knocking the ice when testing various configurations related to polkit and PAM.
I've only recently moved my kernel QA from dusty old busybox to systemd (because I noticed that in some kernel features I was lagging over two years of user space progress), and it is definitely one of the more useful features of systemd for my work.
You can also pretty much use anything you want in the foreseeable future. In the end of the day, all of these are roughly front-ends configuring privileged access based on PAM, PolKit, MAC and DAC in varying levels of detail. There's also some other similar tools such as "really".
0
0
0
@chexum yea very possible! thanks for insight.
was not something i was suspicious, just was wondering why is it required :-) does wayland still use xscreensaver? my desktop knowledge is lacking..
was not something i was suspicious, just was wondering why is it required :-) does wayland still use xscreensaver? my desktop knowledge is lacking..
1
0
1
Jarkko Sakkinen
jarkko
Edited 1 year ago
rpm -qpR signal-desktop-7.13.0.aarch64.rpm
/bin/sh
/bin/sh
at-spi2-core
gtk3
libXScrnSaver
libXtst
libnotify
libuuid
nss
rpmlib(CompressedFileNames) <= 3.0.4-1
rpmlib(PayloadFilesHavePrefix) <= 4.0-1
rpmlib(PayloadIsXz) <= 5.2-1
xdg-utils
I don’t get the #xscreensaver dependency that #Signal has…
2
0
0
@liw yeah i thought first thought that it was a cheap set of random tools either from clas ohlson or discount basket of lidl ;-)
0
0
1
@ikkeT mä haluisin itseasiassa oppia kunnolla quadletit ihan raaka-podmanilla. Parempi investointi pidemmän päälle. tilasin sitä varten ihan kirjankin: "DevOps for the Desperate" ;-)
https://nostarch.com/devops-desperate
https://nostarch.com/devops-desperate
0
0
1
Jarkko Sakkinen
jarkko
Edited 1 year ago
Just an observation but if considering testing ARM64-Linux, Apple is actually a “budget choice” compared to:
- Low-volume Linux ARM64 laptops (few available).
- ARM64 server racks.
Especially if you take -1 gen Apple hardware, which you can get for bargain prices pre/post the next WWDC ;-) The laptop seen in previous screenshot is M1 Pro, and it is quite nice and fast , and that is even -2 gen.
1
3
4
@ikkeT tässä on oleellista on https://asahilinux.org/ sinänsä mulle käy mikä vaan, kunhan on systemd eikä ubuntu :-)
applehan on deskaritason (ja siitä ylöspäin) arm64-raudassa budjettivaihtoehto. muut vaihtoehdot on ylihintaiset linux-laptop-valmistajien tuotteet tai monen tonnin serveriräkki :-)
applehan on deskaritason (ja siitä ylöspäin) arm64-raudassa budjettivaihtoehto. muut vaihtoehdot on ylihintaiset linux-laptop-valmistajien tuotteet tai monen tonnin serveriräkki :-)
0
0
0
Jarkko Sakkinen
jarkko
0
0
0
Jarkko Sakkinen
jarkko
Rick Beato is like Anthony Bourdain of music, favorite tubers: https://www.youtube.com/watch?v=1bZ0OSEViyo
0
0
1
Jarkko Sakkinen
jarkko
#note to myself, not throughly thought out...
Useful #kernel feature for #hibernate might be a file that would tell the number of pages required for hibernate.
The "O-figure" for hibernate is AFAIK O(n+sqrt(n)) but exact number is dependent on kernel internals but in theory could be calculated by simulating the process without committing.
Instead of swap it might also be useful to have "spare" volume of ~100 GB for doing hibernate and e2scrub type of operations, which could have its own design for encryption etc. Swap for swapping overcommitted memory more than few at most GiB is not that useful these days.
#linux
Useful #kernel feature for #hibernate might be a file that would tell the number of pages required for hibernate.
The "O-figure" for hibernate is AFAIK O(n+sqrt(n)) but exact number is dependent on kernel internals but in theory could be calculated by simulating the process without committing.
Instead of swap it might also be useful to have "spare" volume of ~100 GB for doing hibernate and e2scrub type of operations, which could have its own design for encryption etc. Swap for swapping overcommitted memory more than few at most GiB is not that useful these days.
#linux
0
1
1