HN Simulatornew | past | comments | lists | submit | fromlogin

To prevent evil maid attacks you'd need to use verification of the entire boot chain, so bootloader, kernel, and system packages would be e.g. signed and each verifying each subsequent step before executing it.

Encrypting instead of verifying would still allow for evil maid attacks since they can just replace it with something that ask for a key and never really does anything with it other than sending it to a remote server.

To reply to your "And where is the trust store database stored by default?": The root store would be the motherboards secure boot databases for the signed bootloader (or kernel if doing direct boot). the following steps can verify however they like. E.g. something like ParticleOS[0] uses the roothash of the usr-verity partition, which was embedded into the signed UKI.

[0]: https://github.com/systemd/particleos


Nspawn has a lot of issues with bind mounts in general. They've never been fixed because pottery doesn't see the point. See for example: https://github.com/systemd/systemd/issues/26897

I'll give this one a try, thanks for sharing!

FWIW, I'm currently using systemd-nspawn via mkosi: https://github.com/systemd/mkosi

It makes an image and runs it in separate namespace. It can start at bash or init. It's very fast but there seem to be a problem creating an Ubuntu image on Debian and vice versa.


Yeah - and also trying to promote it. I watched this lately.

First, the Anthropic master clown babbles about how mankind will be destroyed by AI. I mean, skynet in the movie Terminator was kind of cool; the current AI models just work by slopness. They want to erode us by dragging us down to zero quality.

Then this continued over the last some days. Today I heard at the UN speech a german minister babble about AI building up nukes in a private garden (a large garden), bla bla bla all the information is there bla bla bla AI mass-nuke generation.

Now, they are clearly FUDding many people who don't understand anything here. People will associate AI with mega-danger. Probably there will be laws that ensure that only a few companies can control the AI slopness soon, because IT IS SO DANGEROUS. We saw the same discussion with "omg omg omg terrorists", then "omg omg omg someone protect the children", e. g. as a rationale for everyone to ID in order to access information on the world wide web (starting with the anti-social websites, such as facebook, but it won't be limited here; the computer will hand over your ID to the government; see systemd preparing for this move: https://github.com/systemd/systemd/pull/40954 - it all comes in small baby steps, until all the final legislation is in place. The same happened before again and again - anyone remembers the DMCA? Corporations lobbied and paid for that - and it happened).


Here's the PR https://github.com/systemd/systemd/pull/40954/changes

It was like 100 lines, a third of it is tests, and another third is a parse_calendar_date function.

What are you talking about, "infrastructure"? How complicated do you think storing a date is?




I do not want my computer to yield information to evil outside actors in general. This is why systemd going that way is so outrageous:

https://github.com/systemd/systemd/pull/40954


You do need some root process to be able to write the updated kernel image to a root owned directory. On macos that is presumably their update daemon, while on something like ParticleOS[0] IIRC it is the systemd-sysupdated daemon, though I don't know if you can initiate the update as an unprivileged user or if its done on a timer or only root initiated (sysupdate has had a lot of changes not too long ago that reworked it extensively. It use to be something only root invoked transiently with a possible timer to "update all", but now it has a full on daemon).

Basically the reason we "need sudo access" (or anything to elevate privs to root) is because how the system is architected and changing that is mostly only possible on new distros which can decide to change the architecture.

[0]: https://github.com/systemd/particleos


So is anyone going to revert the commits that jumped the gun on this?

E.g. systemd's birthdate field https://github.com/systemd/systemd/pull/40954


> Getting media to work on Linux is more complex than you may think.

Not at all: ffmpeg + mpv

Problem solved.

> a lot less feasible for Linux distributions designed to be freely downloaded and redistributed.

> [...] spent some time enabling third-party repositories to get multimedia codecs, you already know what it's like

Well, this was quite easy on debian, but I abandoned being a slave to distributions. I compile from source as-is, using a set of ruby scripts (primarily), not unlike homebrew, but based on YAML rather than ruby classes 1:1 (this YAML dataset is expanded into variables that fully describe how a program is installed, what it installs, and so forth, so any other programming language can also evaluate the same dataset here; I just wrote the first implementation via ruby as such, but I can add support for other languages easily, or someone else can, it is open by design).

What the article describes really is, at worst, a problem for some distribution. But even there it was never any problem to have support for multimedia, unless the distribution sucks. Many distributions suck, but many other distributions are also excellent. Arch is IMO much better than debian, by default. In fact, both debian and the rpm-based distributions, are usually quite retarded by default. Arch, Gentoo and many more, even oldschool slackware, are much saner by default. I have no idea why retarded distributions became so dominant, probably because they were very easy to get going (see ubuntu). But quite frankly, the article really blows this all out of proportion. Enabling support for codecs is easy, even on most of the retarded distributions.

I don't see video patent being a problem though really. The real test step for linux will be the upcoming mandatory age sniffing. People already looked with deep worry at systemd running ahead and supporting age sniffing (see https://github.com/systemd/systemd/pull/40954). Now the legislation is slowly changing to force everyone to give up their age. Microsoft will push next here, via Windows. You heard this here first.

Age sniffing will come - the lobbyists were all already paid to force this. The big commercial distributions will go force and ID its users.


I have been complaining about journald abysmal performances for almost as long as I can remember. Here is my latest documented benchmark from 2023, which was slightly better than the one I ran in 2020.

$ time journalctl > /tmp/all.log

real 1m11.364s user 0m52.299s sys 0m6.540s

$ time wc -l /tmp/all.log 3659597 /tmp/all.log

real 0m0.152s user 0m0.056s sys 0m0.096s

$ time journalctl | grep sshd | wc -l

12944

real 0m53.973s user 0m49.535s sys 0m5.210s

$ time grep sshd /tmp/all.log | wc -l 12944

real 0m0.429s user 0m0.332s sys 0m0.100s

https://github.com/systemd/systemd/issues/2460#issuecomment-...


great and clear summary thank you!

I would laso add that if my design decisions or development actions lead to an issue affecting multiple linux distro defaults I would feel responsible and rush for a solid fix instead of this https://github.com/systemd/systemd/issues/15292#issuecomment...


I think this also explains a lot and should not be ignored. https://github.com/systemd/systemd/issues/15292#issuecomment...

It seems a recurring (handling) issue but unfortunately it affects multiple linux distro defaults. This is the worse that can collaboratively happen for FOSS in general imho.


What does the (currently latest, https://github.com/systemd/systemd/issues/40262#issuecomment...) comment mean? Who are the "large folio people"?

journald does do a rewrite of the log file on rotation, so you're paying that IO anyway even if you ignore the dumb hash table updates.

https://github.com/systemd/systemd/blob/8f4cd7de43d1e6e94687...

WAL writeback is at least principled and efficient. It works out to being equivalent to the custom Parquet-rotation things others mention, but already implemented and working.

So, yes, WAL for logs, because LSM is the design everyone converges on and a WAL is LSM. Better to use the LSM already implemented and debugged in a database than write some random new one in terms of Parquet that's going to have to do the same stuff in the end anyway, just with novel bugs and no tool support.

(And look, I don't give a damn what "DB" people say, a WAL writing back to a DB IS log... structured... merge under any fucking sensible definition of what LSM means.)



it has bad scaling properties esp. if you have many journal files

the last time I contributed to journald upstream was to fix a degenerate behavior with many journal files: https://github.com/systemd/systemd/commit/176f73272e6e3116ca...

that makes a dramatic difference for those hitting this case, but it only gets things from nearly unusable to slow-as-usual.


journald uses hash tables, I think it update it on every new log line, although I didn't debug it in depth yet.

https://github.com/systemd/systemd/blob/199f75205b9c0625bf56...



> A sane person would have made it a Daemon that processes talk to surely?

> Really it should probably be part of SystemD

https://github.com/systemd/systemd/pull/39855


That's good, but the lobbyists in the EU will continue to try to destroy VPN. The "we must protect children" is the ruse they are using right now. Everyone will have a unique ID in the future, otherwise you will be unable to browse (at the least for commercial systems such as Microsoft-based computer systems; I can not yet predict how it will work for Linux but systemd already has that field for your age: https://github.com/systemd/systemd/pull/40954 - Poettering forbid further discussion about that issue too)

Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact

Search: