@derobert Heh, yeah, I've had that happen a few times too. not on a self-answered one, those I usually remember. I just searched and found the U&L link, it solved my problem perfectly, went to upvote and "you can't vote on your own posts"
OK, before I post a question and get yelled at in the comments, can someone explain to me why netbsd uses ftp even though it's apparently not secure ? I thought netbsd had one of the goals as security
@SergiyKolodyazhnyy No, no, no, no!! That should absolutely not happen, and it should most certainly not happen as a function of your rep! Please let me know if it does and I'll slap whoever is complaining.
Eh, I'll try to form a coherent question later and ask on the infosec site. It just kinda bothers me that netbsd claims itself to be secure, yet uses ftp for package management which people agree isn't secure. At least it's the impression I get. But then again, if package is already tampered with on the other end, it doesn't matter if it's ftp or not
Ubuntu,too. It's all http in sources.list, no https, except for a few private packages from launchpad and Skype for linux repo
@SergiyKolodyazhnyy I'm not familiar with exact details on freebsd package management/downloads, but I would expect that they do not require a secure channel any more than any other similar system.
Downloads can (and should) be verified using digital signatures. If the signature is valid, the download is secure in sense it has not been altered.
Should your download have different checksum (for any reason), the package manager would reject it. If the gpg signature verification (for Release file) fails, apt won't install anything from the repository.
It is quite well explained on the linked debian wiki page.
there is a question I posted answers since there is more than one way to solve the problem. There was a 3rd perfectly good answer (not by me) now deleted by answer :\. Should I write an answer covering the 3rd alternative, or should mod un-delete the deleted answer?
Another user asked me in the comments section to my answer here why I combined so many approaches in a single answer. This made me realize we don't seem to have a clear consensus on if and when it is proper to post multiple answers.
I found this thread on the main meta:What is the official etiq...
Personally, I find a single comprehensive answer to be better but that's just me and not my modly persona. So nothing wrong with posting many if you prefer. Just make sure you only do so when the approaches are significantly different.
@terdon well, now that I wrote them all I am not anymore sure about what the OP was asking about. Guess no point of making a close vote since answers cover the possibilities I can think of
@Videonauth I wonder why it is set up like that. Would be a reasonable question, actually. If you feel like asking it. I haven't observed this behavior personally.
> In a systemd world we made this more dynamic: in order to make things more efficient login prompts are now started on demand only. As you switch to the VTs the getty service is instantiated to getty@tty2.service,
@sebasth Looks like a systemd thing. But I'm not seeing that here.
@Videonauth Marginally more efficient.
> This behaviour is mostly transparent to the user: if the user activates a VT the getty is started right-away, so that the user will hardly notice that it wasn't running all the time.
Hmm, maybe it's happening and I'm not noticing it.
in the end reboot using some kernel command line options to systemd masking gdm3 did not work for reason I never cared to investigate
eventually just booted init=/bin/bash and corrected the configuration
eventually when I hit the issue again and went to make some relaxing coffee instead I learned that re spawning does stop eventually, (some minutes)
some heuristics would be in place, should the service file X times in a second or two, it probably won't make a difference in many situations if the restart is tried 10 or 1000 times
@FaheemMitha also wondered why umask doesn't seem to work on gnome3, turned out since gnome uses systemd as user session manager it also means systemd overwrites the umask by its own hardcoded value. Made a bug report back then, I am not holding my breath on it being resolved anytime soon.
in general I kinda do like some stuff systemd can do, such as managing all the processes and sub-processes spawned by a daemon (convenient should you want to get rid all of them and the daemon itself lost account)
Debian has mostly gone along with the systemd thing. I don't think they would do that if they thought it was harmful. But I wonder if it is a good thing that it is taking over more and more functionality. Unix was designed as modular for a reason. Thoughts?
and processed ending up in correctly built environments (environmental variables, security contexts, cgroups, namespaces and so on), which on some (older) systems seemed to be a bit wild
now I got wondering does any environment initialization/sanitation take place with openrc when a daemon is started directly from initscript or is there some additional openrc commands I am not aware of
haven't followed that closely, kinda understand it everyone hates it if new stuff breaks old systems (like the umask not being applied on gnome anymore)
it seems to have rather tight coupling between the components it provides to different tasks
there are multiple levels of modularity, kernel does support loadable modules, even that they do run in kernel space after being loaded (vs microkernel and userspace)
@FaheemMitha well sure wages a high here but in turn you pay in many cases more as people on other countries, but then those people earn less in return too so its equal i guess
Finland is a pretty good place to be by international standards. Even if it not as warm as you would like. Have you done much international travelling?