Today I have spent way too much time handling the https://copy.fail situation #copyfail
-
@alexanderkjall I read that they had waited a month with distributing the PoC and that major distributions were prepared.
@LabanSkoller @alexanderkjall if they say so, they are lying. The distros security list wasn't notified and there was no headsup to Debian outside of the list either. And Ubuntu surely neither, otherwise they wouldn't have just pushed a patched kmod package with the module blacklisted...
-
@alexanderkjall But they say they 'Reported to Linux kernel security team' on 23rd March; shouldn't that have triggered the distros finding out?
@penguin42 It did not, the process is somewhat described here: https://docs.kernel.org/process/security-bugs.html
-
Today I have spent way too much time handling the https://copy.fail situation #copyfail
The persons who discovered it didn't notify the distribution security list, so no patched kernels was available for people to install when they released it.
But they did have time to write an exploit, and thought it was a good idea to distribute that on day one, before vendors had time to provide patches.
I'm not very impressed with xint.io, I guess it's the marketing department that runs the show.
An "AI-assisted finding", using a pretty AI-written and designed looking website, m-dash, semicolons, it has it all
Couldn't their AI also tell them how to report it properly??? -
Today I have spent way too much time handling the https://copy.fail situation #copyfail
The persons who discovered it didn't notify the distribution security list, so no patched kernels was available for people to install when they released it.
But they did have time to write an exploit, and thought it was a good idea to distribute that on day one, before vendors had time to provide patches.
I'm not very impressed with xint.io, I guess it's the marketing department that runs the show.
@alexanderkjall Brad Spender (GRSecurity) has been highly critical of the Linux Kernel security bug handling process since forever, and one of those criticisms is that the members of security@kernel.org don't notify the linux-distros security list, or really triage severity in a way that he approves of as a security vendor and practitioner, their "security bugs are just bugs" stance that refuses to give priority to security issues is infuriating to some people who see security bugs as higher priority than any other kind of bug.
-
@LabanSkoller @alexanderkjall if they say so, they are lying. The distros security list wasn't notified and there was no headsup to Debian outside of the list either. And Ubuntu surely neither, otherwise they wouldn't have just pushed a patched kmod package with the module blacklisted...
@jmm @alexanderkjall I think I mixed it up with the Linux kernel security team. But shouldn’t *that* team notify the distros?
-
@jmm @alexanderkjall I think I mixed it up with the Linux kernel security team. But shouldn’t *that* team notify the distros?
@LabanSkoller @jmm They do not, the process is somewhat described here: https://docs.kernel.org/process/security-bugs.html
-
@penguin42 It did not, the process is somewhat described here: https://docs.kernel.org/process/security-bugs.html
@alexanderkjall That feels too complicated to leave to leave just to a (potentially 1st time) reporter. I would have hoped that at the very least the LK security team would track with the reporter and remind them of the need to do the other bits regularly. Especially on a nasty one!
-
@LabanSkoller @alexanderkjall if they say so, they are lying. The distros security list wasn't notified and there was no headsup to Debian outside of the list either. And Ubuntu surely neither, otherwise they wouldn't have just pushed a patched kmod package with the module blacklisted...
@jmm @LabanSkoller @alexanderkjall yeah, it's definitely not been handled optimally. on the RH side, Fedora happens to be OK as the fix landed upstream in 6.19.12 and that already went stable, but RHEL (and hence CentOS and probably Alma and Rocky) are affected with no day-0 update - https://access.redhat.com/security/cve/cve-2026-31431 .
-
Today I have spent way too much time handling the https://copy.fail situation #copyfail
The persons who discovered it didn't notify the distribution security list, so no patched kernels was available for people to install when they released it.
But they did have time to write an exploit, and thought it was a good idea to distribute that on day one, before vendors had time to provide patches.
I'm not very impressed with xint.io, I guess it's the marketing department that runs the show.
@alexanderkjall And there I sat, thinking it was just me being too dumb to figure out whether I had a patched kernel without running their bespoke, obfuscated script.
-
@alexanderkjall Brad Spender (GRSecurity) has been highly critical of the Linux Kernel security bug handling process since forever, and one of those criticisms is that the members of security@kernel.org don't notify the linux-distros security list, or really triage severity in a way that he approves of as a security vendor and practitioner, their "security bugs are just bugs" stance that refuses to give priority to security issues is infuriating to some people who see security bugs as higher priority than any other kind of bug.
@raven667 @alexanderkjall Sounds like reasonable criticism to me. But then, an extended group of volunteers can only be expected to do so much. If we want SLAs, we have to pay people.
-
@LabanSkoller @jmm They do not, the process is somewhat described here: https://docs.kernel.org/process/security-bugs.html
@alexanderkjall @jmm hmm…
> As such, the kernel security team strongly recommends that as a reporter of a potential security issue you DO NOT contact the “linux-distros” mailing list UNTIL a fix is accepted by the affected code’s maintainers and you have read the distros wiki page above and you fully understand the requirements that contacting “linux-distros” will impose on you and the kernel community.Well, if it’s too complicated to be a reporter, there is always fulldisclosure@seclists.org.
-
@raven667 @alexanderkjall Sounds like reasonable criticism to me. But then, an extended group of volunteers can only be expected to do so much. If we want SLAs, we have to pay people.
@OmegaPolice @alexanderkjall yes and the majority of Linux kernel development is not volunteers, its a consortium of vendors organized through the Linux Foundation trade org which _does_ pay people. There are still volunteers who work on Linux but they shouldn't be a shield for the majority to pretend they are "just a smol bean, uwu" and dont have some responsibilities.
I'm the *first* to say that dragging volunteer FOSS maintainers is shitty (and that volunteers shouldn't cosplay as commercial vendors, by doing things like having public issue trackers, as it is nonsense to "open a support case" with a tracking number against a *volunteer*) but Linux kernel has volunteers it is not a volunteer led project anymore, its a consortium of major companies which benefit from pooling effort and having a neutral forum to collaborate.
I think security focused people, whose customers are also prioritizing security sometimes dont have empathy for or understand people who don't have security as their top priority, and treat people who don't share their priorities as stupid and incompetent, rather than understanding the differences in goals and constraints. That said there is probably room for improvement on kernel security, but more from a better systematic approach to prevent defects than treating every bug with a website as some super secret special thing. the current design makes a constant stream of local exploits inevitable
-
@OmegaPolice @alexanderkjall yes and the majority of Linux kernel development is not volunteers, its a consortium of vendors organized through the Linux Foundation trade org which _does_ pay people. There are still volunteers who work on Linux but they shouldn't be a shield for the majority to pretend they are "just a smol bean, uwu" and dont have some responsibilities.
I'm the *first* to say that dragging volunteer FOSS maintainers is shitty (and that volunteers shouldn't cosplay as commercial vendors, by doing things like having public issue trackers, as it is nonsense to "open a support case" with a tracking number against a *volunteer*) but Linux kernel has volunteers it is not a volunteer led project anymore, its a consortium of major companies which benefit from pooling effort and having a neutral forum to collaborate.
I think security focused people, whose customers are also prioritizing security sometimes dont have empathy for or understand people who don't have security as their top priority, and treat people who don't share their priorities as stupid and incompetent, rather than understanding the differences in goals and constraints. That said there is probably room for improvement on kernel security, but more from a better systematic approach to prevent defects than treating every bug with a website as some super secret special thing. the current design makes a constant stream of local exploits inevitable
@raven667 @OmegaPolice I agree that it would be great if the kernel security team had a process that made life simpler for downstream vendors.
But since neither me or my employer contributes anything to make that happen I don't think it's my place to have public opinions about it.
Personally I would love to see more effort focused on reducing the attack surface of the kernel.
-
@jmm @alexanderkjall I think I mixed it up with the Linux kernel security team. But shouldn’t *that* team notify the distros?
@LabanSkoller @jmm @alexanderkjall No non I actually read that too from the FAQ on the Copyfail page yesterday. So. That was a lie?
-
Today I have spent way too much time handling the https://copy.fail situation #copyfail
The persons who discovered it didn't notify the distribution security list, so no patched kernels was available for people to install when they released it.
But they did have time to write an exploit, and thought it was a good idea to distribute that on day one, before vendors had time to provide patches.
I'm not very impressed with xint.io, I guess it's the marketing department that runs the show.
@alexanderkjall and strangely disclosed just before the 1st of may, like the mongobleed disclosed just before 1st of January.
Almost as a buzz strategy, pushing IT folks to work on weekends and public holidays.
Seriously, waiting next Monday, letting a weekend to all distros was so hard? -
@alexanderkjall That feels too complicated to leave to leave just to a (potentially 1st time) reporter. I would have hoped that at the very least the LK security team would track with the reporter and remind them of the need to do the other bits regularly. Especially on a nasty one!
@penguin42 @alexanderkjall I agree, having to report to 3 different lists, in a particular order, all with their own policies and methods of working, seems overly complex.
(the complexity may be necessary, but I can see why someone might miss out some steps)
-
Today I have spent way too much time handling the https://copy.fail situation #copyfail
The persons who discovered it didn't notify the distribution security list, so no patched kernels was available for people to install when they released it.
But they did have time to write an exploit, and thought it was a good idea to distribute that on day one, before vendors had time to provide patches.
I'm not very impressed with xint.io, I guess it's the marketing department that runs the show.
That's not what the disclosure timeline claims:
2026-03-23 Reported to Linux kernel security team
2026-03-24 Initial acknowledgment
2026-03-25 Patches proposed and reviewed
2026-04-01 Patch committed to mainline
2026-04-22 CVE-2026-31431 assigned
2026-04-29 Public disclosure (https://copy.fail/)Is this timeline in error?
-
@alexanderkjall That feels too complicated to leave to leave just to a (potentially 1st time) reporter. I would have hoped that at the very least the LK security team would track with the reporter and remind them of the need to do the other bits regularly. Especially on a nasty one!
@penguin42 I place some amount of blame with distro maintainers for not following up on patches released by the kernel team. I know it's not a thankful job, but it needs to be done.
-
@penguin42 I place some amount of blame with distro maintainers for not following up on patches released by the kernel team. I know it's not a thankful job, but it needs to be done.
@fedops @alexanderkjall In this case it doesn't seem to have been that simple; The backports to older kernels were only released upstream yesterday, and the CVE only got assigned a week or two back; so it's not clear if anyone new. I'd love to know if the kernel security guys told anyone there were some pending.
-
@fedops @alexanderkjall In this case it doesn't seem to have been that simple; The backports to older kernels were only released upstream yesterday, and the CVE only got assigned a week or two back; so it's not clear if anyone new. I'd love to know if the kernel security guys told anyone there were some pending.
@penguin42 yeah I understand the back ports problem but there was a patch in mainlinel. The timeline says:
2026-03-25 Patches proposed and reviewed
2026-04-01 Patch committed to mainlineSurely "proposed and reviewed" must have caught someone's interest?
The sad thing is a patch wasn't even necessary. Disabling the ALG function via bootparam was completely enough except in rare cases where crypto performance actually counts.
