jude-

68 ms·

571.
by jude- 12y ago
> When the traditional sysvinit starts and stops processes, it actually has no clue what it's doing and thus init scripts need to rely on PID files and other hacks to provide basic functionality. Tooting my own horn here, but I wrot
572.
by jude- 12y ago
Sorry, I should have been clearer. I was referring to the cost/performance trade-offs when building a storage tier. The choice of persistent storage for a laptop obviously presents different trade-offs.
573.
by jude- 12y ago
I doubt it. If your access patterns follow a power law distribution, why should you pay more for less space, if the performance boost gained by doing so (i.e. putting the "long tail" onto fast storage) is negligible? Moreover, t
574.
by jude- 12y ago
I'm not seeing how this is any different from true directories in terms of time and space complexity, as far as rename and path resolution are concerned. Sure, the trie nodes don't have to be implemented as key/value pairs i
575.
by jude- 12y ago
I'd be curious to know how to make it more efficient. Either the descendants of a directory include the directory's name in their identifier, or they don't. If they do, then to implement rename() you need to do something su
576.
by jude- 12y ago
> Is there a real fundamental difference between the two approaches? Yes. With true directories, you can implement directory rename() with O(a+b) time complexity, where a and b are the depths of the paths to the directories. With a fla
577.
by jude- 12y ago
I took a quick look. I have some doubts about your anti-Sybil attack measures: * The more CPU my arbiter nodes can prove that I have, the heavier my arbiter vote. But at the same time, I can pretend to be many, many small nodes, which I&#x
578.
by jude- 12y ago
I think creating a purpose-specific overlay network that gives you performance comparable to what we see today with centralized solutions while simultaneously preserving the guarantees from the blockchain is the hardest problem here by far
579.
by jude- 12y ago
While I agree with Theo in principle, it sounds like he's precluding the possibility that one can create a virtualization layer with formally-proven correctness guarantees (be it a hypervisor, an OS, or an application VM). For example
580.
by jude- 12y ago
> whereas dbus and systemd and gstreamer can move fast and break things I think you will find that this attitude is exactly the opposite of what is expected at the plumbing layer. Move fast and break things is fine when no layers depend
581.
by jude- 12y ago
FWIW, their threat model assumes that Alice can't subvert the software with hardware-based attacks. Relying on this system for DRM would be pretty shortsighted, since (1) Alice obviously has physical access to her hardware, and (2) Al
582.
by jude- 12y ago
And if you run gitstats on systemd, you'll see that just 10 people are responsible for over 90% of the code.
583.
by jude- 12y ago
> Service A starts writes PID to disk, lets say 123. lots of processes start and stop as it goes along and does its work. Service A crashes/stops working PID 123 gets reused by a new process SysAdmin comes and hits /etc/in
584.
by jude- 12y ago
Exactly. I'm (obliquely) affirming your argument that the commonly thrown-around excuse of "you don't need systemd, you just need something that implements its DBus interfaces" is a distinction without a difference.
585.
by jude- 12y ago
That's the fundamental problem with DBus-exposed interfaces: they're APIs that get marketed as abstractions. APIs don't leave much room for different interpretations, meaning you'll end up rewriting the component you&#
586.
by jude- 12y ago
> If they feel like learning systemd is a chore, they might be in the wrong business. I think IT managers would prefer it if they didn't have to spend time and money re-training their sysadmins or hiring/firing them to ensure
587.
by jude- 12y ago
One thing that needs further investigation is the degree to which distros adopted systemd because they'd rather not maintain patches for udev, logind, KDE, GNOME, Wayland, etc. to keep them decoupled. I have a hunch that for !RedHat, t
588.
by jude- 12y ago
I'm pretty sure it's Greg KH who's running kdbus these days [0]. At least, he's the one who submitted the patch to lkml [1]. [0] https://github.com/gregkh/kdbus [1] https://lkml.org&#x2F
589.
by jude- 12y ago
> you dont have to use journald. you can send back from journald to syslog That's a contradiction. You're running journald even if it's dead weight.
590.
by jude- 12y ago
Either I'm using journald to log to a binary format, or I'm having journald forward everything to syslog. If I'm trying to avoid binary logs in the first place, then there's really no point for journald to be running--j
591.
by jude- 12y ago
Tolerating intolerance is not an act of love, but of apathy--both towards the bigot, and the bigot's targets.
592.
by jude- 12y ago
Fingerprints identify individuals, but they should not be relied upon for authentication (which is what the article was getting at). Unlike a password or private key, you can't change your fingerprints when they get compromised.
593.
by jude- 12y ago
It's my understanding that it's not fixable without some help from the kernel. The fundamental problem is that a program can race ahead of a userspace syscall policy enforcement framework (i.e. sysjail) by trapping to the kernel
594.
by jude- 12y ago
> OpenRC isn't really a replacement for sysvinit; rather it bolts on some functionality on top of sysvinit. My understanding is that OpenRC is not an incredibly mature codebase and would have needed a lot of work to come up to featu
595.
by jude- 12y ago
Thank you for taking the time to write this up! (the link is also appreciated :) I think a lot of the rage and loathing regarding systemd is due to its coupling to the rest of the OS. I think a lot of !systemd users were quiet on this init
596.
by jude- 12y ago
All of systemd's other components require systemd-PID-1. So yes, systemd does usurp PID 1.
597.
by jude- 12y ago
> Tagged binary formats are easily and trivially self-describing, and as resistant to corruption as ASCII. The advantage of human-readable logs is that the semantics and rules of the language it's written in serve as a built-in forw
598.
by jude- 12y ago
sysjail is not safe for jail-like use. Sysjail'ed processes can do Bad Things by exploiting concurrency races in the syscall interception layer. Relevant paper: http://www.watson.org/~robert/2007woot/2007usen
599.
by jude- 12y ago
I sincerely hope you are joking. Dbus-as-IPC doesn't make for composability at all, since each program must be aware of (1) the names and signatures of each other program's dbus-accessible methods, and (2) each other program&#x27
600.
by jude- 12y ago
> The thing is we're already dealing with disjoint sets of programs and broken composability. Thats why so many shell one liners take the form...I mean we have to do the C->Bash->Awk->Bash->C precisely because nothing ac

More ›