It looks like the expected thing happened.

The kernel devs patched the kernel. The kernel devs have a pretty known, straightforward stance in how they ship fixes for anything, because anything in the kernel can be a security problem.

Distro maintainers can see kernel changes. Some distros aggressively track new changes. Others backport what they feel are relevant. Others don’t do either.

Users pick what distro they use, and how they set up their infra.

Maybe if I were paying for RHEL licenses I’d be eyeballing the money I pay and RHEL’s response time.

But the ownership here lies with system operators, who pick their infrastructure, who design their security model, and who build their operational workflows. This vuln is a great example: people who looked at shared untrusted workloads on a single kernel and said “Hell no” had a much calmer day than teams who thought that was a good idea.

▲

SOLAR_FIELDS 2 hours ago | parent [-]

The fact that you had to take a whole paragraph to explain the contortionist arrival at something that isn't even really super clear after you explained it (you kinda pointed the finger both at end users and at distro maintainers simultaneously) and essentially boils down to "well, you as the end user need to be following kernel CVE's and can't trust distro maintainers to do it" does in fact indicate that there is a deeper issue at play here. You might say "well, there's no implicit chain of trust here". You might be right, but is that really the most effective way of doing things? Of course Linux is Use it at your Own Risk, but is there not a concept of "we as a collective community should get together and try not to drop the ball on some serious shit?"

In terms of something actionable, and maybe someone more well versed in how the distros work can tell me why this is a bad idea, but shouldn't there be a documented process and channel for critical CVE's to be bubbled out to distro maintainers who then have some sort of SLA for patching them and sending them downstream to end users? Perhaps incentives are not aligned to produce this outcome.

▲

PearlRiver 9 minutes ago | parent | next [-]

The real advantage of Microsoft is that there is someone you can sue!

Linux like every open source project is just a bunch of people who are YOLOing it. Not something you use for your fortune 500 critical mission infrastructure.

▲

akerl_ an hour ago | parent | prev [-]

To be more blunt: if you’re paying for a product, the vendor owes you whatever things they committed to. If you’re a Redhat customer and your agreed SLA with Redhat for this kind of security fix was passed by, go be mad at Redhat. (I don’t think Redhat is bad here, they’re just the vendor most known for a commercial offering from the lists here. I would say the same thing about Ubuntu Pro)

Otherwise, it’s on the end user. Distro volunteers don’t owe you anything. Kernel devs don’t owe you anything.

I don’t care about what would be the most effective way of doing things. I care about what folks involved actually owe to each other, and distro volunteers don’t owe users any kind of active chasing of remediation due to the user’s threat model.

The idea of making some kind of streamlined process that solves what you didn’t like about this vulnerability’s remediation is that it ignores basically all the complexity. Like “what about distros that don’t abide by embargoes” or “what distros count as ones that matter” or “what about all the vulns that aren’t in Linux, they’re in software that’s packaged across many operating systems”.

▲

SOLAR_FIELDS an hour ago | parent | next [-]

Right, you’re saying “system is working as designed”, and I’m agreeing, but I’m saying “the system as designed kind of sucks, how can we make it better”?

	▲	akerl_ an hour ago \| parent \| next [-]
		I disagree that it sucks. It leverages a ton of people putting in their time and resources, and relies on system operators being active participants. This vulnerability is, for some threat models, a really big deal. A security group found the vulnerability. They disclosed it. It was patched. Folks here have gotten all kinds of bent out of shape that the groups involved didnt do things in the way each internet commenter would have liked. But this is the system working.
	▲	tptacek an hour ago \| parent \| prev [-]
		Start a distro with your preferred upstream tracking policy.

▲

komali2 27 minutes ago | parent | prev [-]

Just as a purely intellectual exercise, what changes about this if we leave aside ideas of "owe," "deserve ," and "earn?"

There's not really an enforcement mechanism in FOSS like there is in capitalism world, it just comes down to what we want our part of the world to look like. So I think we'd think more clearly if we leave aside the ideas like "who owes who what." I think it's fun to imagine what sort of motivations and incentives there are if we put away the money ones.