4 ms·

I don't mean to imply that if OpenBSD ran those specific services by default that it would be much worse. My point is that: a) Generally for a server people a

I don't mean to imply that if OpenBSD ran those specific services by default that it would be much worse. My point is that: a) Generally for a server people are running some kind of services, and that's going to include third party software. Not always, but often, and that software is not audited by OBSD team. Even say adding a module to httpd may allow for remote compromise and b) That the tools OpenBSD provides for dealing with an attacker who has managed to gain remote access are insufficient compared to what is available, and especially insufficient for an OS that claims to be focused on security.

camgunz 4y ago
> Generally for a server people are running some kind of services, and that's going to include third party software. Oh so you're saying that things like ASLR, chrooting, and safe programming practices aren't useful in things like httpd, because you might run an insecure PHP app, and the only thing that helps you then is something like SELinux? I think that's pretty wrong. Running services in a chroot with pledge gives you a lot of security. And the other security features [0] (empirically) help reduce the occurrence of remote vulns overall with far less complexity than SELinux (which also totally weirds your system). [0]: https://www.openbsd.org/innovations.html https://www.openbsd.org/innovations.html
- Zurrrrr 4y ago
  > Oh so you're saying that things like ASLR, chrooting, and safe programming practices aren't useful in things like httpd, because you might run an insecure PHP app, and the only thing that helps you then is something like SELinux? You sure put in a lot of effort to misconstruing me there. So many people in this thread who are responding emotionally and without the knowledge to really have a proper discussion. Instead of trying to defend pledge, maybe do some more reading on SELinux and understand just how fundamentally different the approaches are, and why pledge just can't compare.
  - camgunz 4y ago
    Sorry, tone is hard over the internet. I've been trying to engage in good faith, but now I'm gonna zoom out a little and say you have a lot of dismissive, low to no-content replies all over this thread, and your reply to me here is a good example. I'm honestly trying to understand your argument! But you didn't help me out even a little bit. > You sure put in a lot of effort to misconstruing me there. I assure you I did not, because I was on the toilet (there's a > 50% chance that this is the case with me on HN). > without the knowledge to really have a proper discussion This is pretty arrogant and dismissive; you don't have a great way to gauge my knowledge on this other than a no-true-scotsman on "if you knew what I knew about SELinux, you'd agree with me". > Instead of trying to defend pledge, maybe do some more reading on SELinux and understand just how fundamentally different the approaches are, and why pledge just can't compare. I'm not really defending pledge, and I'm not offended by the prospect that SELinux might provide superior security. I'm just trying to have a discussion with you. I will suggest that there may be other axes on which to evaluate a security solution. For example, maybe I don't want the complexity of SELinux (it is, by any measure, complex) and OpenBSD's security lets me address my threat model while avoiding SELinux' complexity. Or in other words, maybe I don't care that I can't restrict nginx to only bind to a specific port, I just don't want it serving up arbitrary files out of my filesystems: chroot to the rescue. Here are some quotes from your other posts: > lol, ok. You really think root can't access the devices? lol. Super dismissive! > If someone has remote root you are telling me they can't unhide a filesystem? If you know about a bug in OpenBSD's chroot that lets root access any filesystem you should report it. Otherwise, you're making it up and it's not a good argument. You might be thinking, "Hrmph, I'm just sure it's possible because of the power of root", but I might also be thinking, "Hrmph, I'm just sure there's tons of junk bugs in the monstrous SELinux code base, not to mention unintended interactions with features and other subsystems", which is why we need evidence, of which your posts have none. > It's ["unhiding" a filesystem as root] absolutely possible. No idea about a proof of concept, but I don't even think it matters, honestly. Yeah, here we go. > It's cute that the OS might play some tricks to hide them from you, but at the end of the day, that's all it is doing, is hiding. Security by obscurity. "cute" and "play some tricks" are dismissive. Again, if you've got a way to jailbreak OpenBSD's chroot you should report it. It's not security by obscurity, you can look up exactly how it works because OpenBSD is FOSS. > lol, you clearly have no clue. Also super dismissive. > pledge and unveil are toys by comparison "toys" is dismissive. --- Multiple people have tried to engage with you to try to discuss SELinux vs. OpenBSD's other security features, and your replies have uniformly been, "do some more reading on SELinux", only more rude. If your goal here is to make friends and convince people of the merits of SELinux, my guess is you're actively losing ground with this approach.

camgunz 4y ago

> Generally for a server people are running some kind of services, and that's going to include third party software. Oh so you're saying that things like ASLR, chrooting, and safe programming practices aren't useful in things like httpd, because you might run an insecure PHP app, and the only thing that helps you then is something like SELinux? I think that's pretty wrong. Running services in a chroot with pledge gives you a lot of security. And the other security features [0] (empirically) help reduce the occurrence of remote vulns overall with far less complexity than SELinux (which also totally weirds your system). [0]: https://www.openbsd.org/innovations.html https://www.openbsd.org/innovations.html

Zurrrrr 4y ago
> Oh so you're saying that things like ASLR, chrooting, and safe programming practices aren't useful in things like httpd, because you might run an insecure PHP app, and the only thing that helps you then is something like SELinux? You sure put in a lot of effort to misconstruing me there. So many people in this thread who are responding emotionally and without the knowledge to really have a proper discussion. Instead of trying to defend pledge, maybe do some more reading on SELinux and understand just how fundamentally different the approaches are, and why pledge just can't compare.
- camgunz 4y ago
  Sorry, tone is hard over the internet. I've been trying to engage in good faith, but now I'm gonna zoom out a little and say you have a lot of dismissive, low to no-content replies all over this thread, and your reply to me here is a good example. I'm honestly trying to understand your argument! But you didn't help me out even a little bit. > You sure put in a lot of effort to misconstruing me there. I assure you I did not, because I was on the toilet (there's a > 50% chance that this is the case with me on HN). > without the knowledge to really have a proper discussion This is pretty arrogant and dismissive; you don't have a great way to gauge my knowledge on this other than a no-true-scotsman on "if you knew what I knew about SELinux, you'd agree with me". > Instead of trying to defend pledge, maybe do some more reading on SELinux and understand just how fundamentally different the approaches are, and why pledge just can't compare. I'm not really defending pledge, and I'm not offended by the prospect that SELinux might provide superior security. I'm just trying to have a discussion with you. I will suggest that there may be other axes on which to evaluate a security solution. For example, maybe I don't want the complexity of SELinux (it is, by any measure, complex) and OpenBSD's security lets me address my threat model while avoiding SELinux' complexity. Or in other words, maybe I don't care that I can't restrict nginx to only bind to a specific port, I just don't want it serving up arbitrary files out of my filesystems: chroot to the rescue. Here are some quotes from your other posts: > lol, ok. You really think root can't access the devices? lol. Super dismissive! > If someone has remote root you are telling me they can't unhide a filesystem? If you know about a bug in OpenBSD's chroot that lets root access any filesystem you should report it. Otherwise, you're making it up and it's not a good argument. You might be thinking, "Hrmph, I'm just sure it's possible because of the power of root", but I might also be thinking, "Hrmph, I'm just sure there's tons of junk bugs in the monstrous SELinux code base, not to mention unintended interactions with features and other subsystems", which is why we need evidence, of which your posts have none. > It's ["unhiding" a filesystem as root] absolutely possible. No idea about a proof of concept, but I don't even think it matters, honestly. Yeah, here we go. > It's cute that the OS might play some tricks to hide them from you, but at the end of the day, that's all it is doing, is hiding. Security by obscurity. "cute" and "play some tricks" are dismissive. Again, if you've got a way to jailbreak OpenBSD's chroot you should report it. It's not security by obscurity, you can look up exactly how it works because OpenBSD is FOSS. > lol, you clearly have no clue. Also super dismissive. > pledge and unveil are toys by comparison "toys" is dismissive. --- Multiple people have tried to engage with you to try to discuss SELinux vs. OpenBSD's other security features, and your replies have uniformly been, "do some more reading on SELinux", only more rude. If your goal here is to make friends and convince people of the merits of SELinux, my guess is you're actively losing ground with this approach.