> Upgrading to GitLab 19.3.6 on October 22 didn’t fix the vulnerability. The fix isn’t in GitLab’s own code but in a patch to Ruby, which GitLab applies only in its official packages. Our image used standard Ruby, so it didn’t get the patch.
> The bug is in Ruby’s regular expression engine, so any program that compiles regular expressions from untrusted input is exposed, not only GitLab.
I feel like this era of constant exploits/patches is really bad for those who build from source or make their own images vs those who rely on provided images and upgrade them often. I really can't think of a decent solution for those people other than being plugged in 24/7 to the latest news which is really unfortunate and must be exhausting...
At this point if you are self hosting you should have everything behind a VPN. Only companies with their own security and development teams should be putting things on the public internet.
Gosh I'm at a big company and I'm pretty sure all the self-hosted tools are only accessible through VPN-like tech anyway. The only truly public stuff is Jira, Confluence (which I assume it more to do with Atlassian's pricing of self-hosted) and Sharepoint.
Atlassian products (Jira Server) are no longer supported, and their more Enterprise offering dies in 2029, so everything is ending up in their cloud now yeah.
I know a lot of companies vibe-coding alternatives. I'm not an expert but it's perhaps even illegal for most European companies to migrate any customer data to Atlassian servers.
If someone wants to vibe code an alternative to a SaaS tool, my gut feeling is "no, this is bad and it will blow up" as vibe-coded projects tend to do.
But then the alternative is often paying $X million per year for licenses to a half-assed program that hasn't had a UI update since 2005.
I just don't have a good argument on why they shouldn't when the outcome will be just as dogshit, only cheaper and flashier.
They just started updating the UI and I am pissed off.
I don’t want them to change UI. I have plenty of work to do and have plenty of other tools that I use daily.
I have my workflows, I am swamped with work and now I have to change my workflows because someone changed options in UI where ones that O used are not there anymore.
The argument I'd make isn't about UI or features, it's about who owns security afterwards. The vendor has people whose whole job is patching things like this Ruby regex issue. The internal vibe-coded Jira usually has nobody, and the bugs that bite are boring ones, like a project page that loads for anyone who guesses the ID.
If they go ahead anyway, the ask is just that someone actually owns it after launch, not only builds it.
Atlassian is a US-parented company and this means that CLOUD act can affect it. Even this only can make it unsuitable. Then there are questions to data-in-transit. Also nothing is legally tested. Vibe-coding an alternative is much safer if you have sensitive customer data (even simple support tickets).
This person is just posting absolutely baseless claims, as if Atlassian, a company that basically exists purely on the back of B2B SaaS sales, has not considered EU data residency requirements in their core product offering: https://www.atlassian.com/software/data-residency
FWIW nearly every SaaS or IaaS is making bold claims about supporting data sovereignty, but unless it has been challenged in court I wouldn’t put much stock into it.
The fact that Jira/Confluence are being processed by their LLMs puts them in direct legal conflict with several statutes in the EU, its a matter of time for them to get hit by this and I’m betting that they think the fine will be lower than the revenue they acquire for doing it- and they’re hoping they don’t get hit at all.
This seems to be the common case for a lot of the non-EU tech companies, and it seems to be working.
Oracle is currently the market leader in “sovereign” compute for EU governments for example, as ironic as that is.
> This person is just posting absolutely baseless claims
You don't need to be so hostile. I know about the EU requirements but many EU companies have sensitive data on hosted JIRA / Confluence that simply cannot be in the possession of a non-European company. I know because I'm working for one. I cannot even remote-work from outside Europe. Some stuff cannot even leave the VPN. There's also the CLOUD act.
Yeah I thought Jira at least was all built on AWS these days. I've never set up a Jira Cloud instance but I'd be surprised if you don't chose where your data lives.
In short gdpr compliant means you can configure it but it must not be compliant by default and also there is no guarantee.
They are constantly changing their software and settings. More than once they believed that something is legal but can be risky for the atlassian customer.
For example training AI on metadata which they believe are anonymized but are maybe not enough changed.
Training data or metadata permisson can not be configured on all subscriptions. But as a data processor they must follow the contract otherwise they become data controller.
Also there are risks associated with privacy data framework, cloud act etc. to be non compliant just by data transfers to other countries.
> We have not requested CVE identifiers in Ruby for these fixes, at least for now.
> Ruby has generally considered it unsafe to compile and use a regular expression created from untrusted input. We stated this in the advisory for CVE-2022-28738, where we nevertheless treated a double free in Regexp compilation as a vulnerability after a comprehensive assessment. We did the same for CVE-2024-27282.
It's a Bad Idea (TM) to compile a regular expression from an untrusted input.
Despite what anyone would try to sell you, a regular expression is a programming language. It's Always A Bad Idea (TM) to compile untrusted input into any kind of programming language. Don't string concatenate SQL queries, eval is evil, it's all the same idea.
Despite this Ruby should have requested CVE numbers. They created a patch that was available to install and prevented this attack. CVE numbers would have triggered automations to update Ruby in everyones systems.
A regex might be a programming language, a proper regular expression (which there are libraries for, like Google RE2) is just a regular language and I don't think you can even charitably call that a programming language.
What’s interesting with steam boilers and other types of OSHA related injuries is that it’s not enough that the person who’s doing the stupid thing dies: you have to enforce it even more and even higher. People won’t value their life.
We don't do rolling upgrades on steam boilers nor do we do it on our door locks. Nothing is perfect, we just need good enough and we are currently far away from good at all.
When it comes to software it hasn't always been the case. Before the Internet software was considered something stable and even now there are areas, like functional safety critical software, where this is true to some degree.
It's not that it can't be done.
It used to be expensive to a certain extent it will stay expensive, but modern languages, tooling and formal methods will hopefully mitigate that enough.
The world has changed but 15 years ago we thought it was utterly insane to just pull whatever $latest from $internetsite.probablytrusted.org as part of your build and deployment process.
> I really can't think of a decent solution for those people other than being plugged in 24/7 to the latest news which is really unfortunate and must be exhausting
The future is near constant burning of tokens in your software factory to do this automatically. The labs create the increase in demand for their own product.
Turns out building the world on an infrastructure of lies has finally caught up with us.
The way LLMs are developed now is but one possible future we could have experienced. Imagine the future where the LLMs were built by a few governments with black budgets and hidden until a war or some other large event broke out. Then suddenly most of the computers and digital infrastructure shits itself as everything goes wonky.
At least this way we get to see that a shared global infrastructure built on the idea that security is someone else's problem was a really bad idea and we might have enough time to fix it or pull enough plugs we don't go back to the dark age.
That's a for now problem. The future is probably less software surface area exposed to the internet, the removal of unsafe languages from software exposed to the network, services like Cloudflare deploying firewalls to detect exploit attempts, and services that can keep a live database of currently active bots and residential proxies to IP block them.
Alongside the current scanning and patching of software.
I don't disagree. Also probably more logs and more metrics since LLMs can scale instrumentation, signals and auditing far beyond what humans could ever process, which in turn, more tokens. Probably larger Datadog bills too.
The patch for Ruby came out 6 days ago. Most Rails apps do not use the Ruby from the package manager but have a hardcoded version in the repo. So unless you have developers releasing a PR to update ruby the moment it comes out, you aren't safe.
This might feel like a "Yeah obviously" moment, but the fact is that very few Rails apps are running the latest ruby release the moment it comes out.
> build from source or make their own images vs those who rely on provided images and upgrade them often.
I mean - this incident brings it into sharper focus - but it's always been thus.
As a Go developer I often tell people "Be wary of vendoring, you are creating a cache of dependencies that need to be managed and maintained"
Or people that "pin" dependencies, they face the same difficulty.
Why not just use "official" (in this case) images - sometimes they lag too - hell, most Linux distributions lag so much with Go compiler packages that, again, I routinely tell people to avoid them in favour of watching the Go official page
I've been calling this out for forever now and have been repeatedly told that actually, competent devs with C are fine and the C software we use has been battle tested so it's largely secure.
Sorry, it's you. You basically told people that they should expect bugs in old code and shot down the proposed fix without offering any other solution; that's non-constructive by definition.
Xe Iaso's blog posts poking fun at C(++) titled '"No way to prevent this" say users of only language where this regularly happens' has been quite busy this month.
Getting rid of C isn't the holy grail it might appear to be (Ubuntu has been working hard at rewriting sudo and coreutils in Rust and they have had plenty of bugs in the process), but I do hope the C projects will preemptively start fuzzing/running fil-C/invent new safety mechanisms for old code.
And how does Rust help? As far as I know, an integer overflow in Rust panics in debug but just wraps in release (a bad choice in my opinion). Would it help specifically in the code execution part of the exploit?
The integer overflow issue is only part of the problem. There are a couple underlying issues that make this exploit feasible: certain regexes produce extremely long bytecode, the integer overflow issue, and finally, a double free.
Rust will not protect you against exploding regexes, it will not protect you against integer over/underflows (unless you specifically call safe wrapping methods), but it will protect you against double frees if you use it sensibly (nothing prevents you from breaking the borrow checker and messing up memory if you write an unsafe {} block).
The key difference is even If an Integer overflows in Rust It still doesn't violate memory safety in safe Rust and therefore doesn't lead to exploitation.
Depends on what kind of exploitation you're talking about -- integer overflows can still cause out of memory conditions and crash the program, which is a denial-of-service vulnerability when talking about webapps.
But yes, memory-save languages like Rust are a massive help protecting against remote code execution. (unless said memory-safe language introduces its own category of RCE vulnerabilities, e.g. deserialization in Java)
> If you use our alpinelinux/gitlab container image, pull alpinelinux/gitlab:19.3.6 again. The image published on October 27 contains Ruby 3.3.13. Any 19.3.6 image you pulled before October 27, and all older tags, are vulnerable.
shouldn't they tag a new version instead of re-releasing a fixed 19.3.6, if earlier 19.3.6 are vulnerable?
Image tags are not are versions and shouldn't be treated as such. The tag "19.3.6" refers to the version of Gitlab, which hasn't changed. This is why everyone should pin their images to a specific sha and setup a process to get notified on changes
This is a fantastic write up on the security incident. Almost any incident like this will have unknowns. It's good to make it explicit. I especially liked the part of what would have been leaked if the attacker was successful. Maybe they could include what type of hash was used.
If Alpine prides themselves on focusing so much on security, they probably shouldn’t use GitLab - the attack surface is too great and it’s never going to be easy to keep safe and up to date fully, unless you have FTEs whose main job is just maintaining it.
Just get the source code of Forgejo, audit it thoroughly and run an instance of that.
Thinking that Forgejo is a viable replacement for someone who uses GitLab is wild. It's a bit like telling someone to use cgit instead of Forgejo. (yes I know Alpine uses cgit as well)
With the eMudhra situation we saw e.g. Go doing an emergency release that blacklists the certs, while they aren’t shipped with Go, as a protection of dependencies that haven’t caught up.
In this situation, wouldn’t the more responsible behaviour of gitlab be to blacklist Ruby runtimes that enable the CVE?
pretty nice to see that despite suspecting it wasnt really successful, a rebuild and thorough process was followed. lack of evidence doesnt mean nothing happened. better safe than sorry. classy response, hats off!
Does anyone know for sure which unsupported versions of ruby are affected? Best I can tell 3 of the four affect old versions of ruby that are not getting a security patch for this. Update your ruby!
The nasty part is that a version scan would have called that 19.3.6 image fixed. The CVE is filed against GitLab, and 19.3.6 is past the fixed version. The actual fix is in Ruby, and Ruby shipped it without requesting a CVE, so anything scanning the Ruby in the image has no advisory to match. If you build GitLab yourself, you had to read GitLab's release notes and then check which Ruby you were on.
Have researchers coined the term `Superbug` or `MegaVuln` yet?
Surely we're approaching a point where an LLM discovers a critical, unfixable vulnerability in every device running any version of $OS, then irresponsibly discloses it, triggering mass collective trauma among admins worldwide, significant downtime in critical infrastructure and scores of deaths.
We've adopted a zero-risk posture at our org. We've taken down numerous amounts of services over the last few weeks related to PHP, Node, and Python until things are properly vetted.
We joked in our last internal meeting that this is the end of public software... I have a feeling we may be right, at least in the very short term.
That's what Meltdown, Spectre, Rowhammer, and other hardware bugs have been. While you can't patch the hardware, there have always been software level mitigations.
Thankfully a lot of the low level stuff is just not that exposed to the network so even if you find a bug in some firmware, you still have to find a way to access it over the network.
> The bug is in Ruby’s regular expression engine, so any program that compiles regular expressions from untrusted input is exposed, not only GitLab.
I feel like this era of constant exploits/patches is really bad for those who build from source or make their own images vs those who rely on provided images and upgrade them often. I really can't think of a decent solution for those people other than being plugged in 24/7 to the latest news which is really unfortunate and must be exhausting...
If someone wants to vibe code an alternative to a SaaS tool, my gut feeling is "no, this is bad and it will blow up" as vibe-coded projects tend to do.
But then the alternative is often paying $X million per year for licenses to a half-assed program that hasn't had a UI update since 2005.
I just don't have a good argument on why they shouldn't when the outcome will be just as dogshit, only cheaper and flashier.
I don’t want them to change UI. I have plenty of work to do and have plenty of other tools that I use daily.
I have my workflows, I am swamped with work and now I have to change my workflows because someone changed options in UI where ones that O used are not there anymore.
If they go ahead anyway, the ask is just that someone actually owns it after launch, not only builds it.
The fact that Jira/Confluence are being processed by their LLMs puts them in direct legal conflict with several statutes in the EU, its a matter of time for them to get hit by this and I’m betting that they think the fine will be lower than the revenue they acquire for doing it- and they’re hoping they don’t get hit at all.
This seems to be the common case for a lot of the non-EU tech companies, and it seems to be working.
Oracle is currently the market leader in “sovereign” compute for EU governments for example, as ironic as that is.
You don't need to be so hostile. I know about the EU requirements but many EU companies have sensitive data on hosted JIRA / Confluence that simply cannot be in the possession of a non-European company. I know because I'm working for one. I cannot even remote-work from outside Europe. Some stuff cannot even leave the VPN. There's also the CLOUD act.
They are constantly changing their software and settings. More than once they believed that something is legal but can be risky for the atlassian customer.
For example training AI on metadata which they believe are anonymized but are maybe not enough changed.
Training data or metadata permisson can not be configured on all subscriptions. But as a data processor they must follow the contract otherwise they become data controller.
Also there are risks associated with privacy data framework, cloud act etc. to be non compliant just by data transfers to other countries.
I’m doing my part!
https://www.ruby-lang.org/en/news/2026/10/23/regexp-fixes-fo...
> We have not requested CVE identifiers in Ruby for these fixes, at least for now.
> Ruby has generally considered it unsafe to compile and use a regular expression created from untrusted input. We stated this in the advisory for CVE-2022-28738, where we nevertheless treated a double free in Regexp compilation as a vulnerability after a comprehensive assessment. We did the same for CVE-2024-27282.
Despite what anyone would try to sell you, a regular expression is a programming language. It's Always A Bad Idea (TM) to compile untrusted input into any kind of programming language. Don't string concatenate SQL queries, eval is evil, it's all the same idea.
i agree with you!
nevertheless,
> where we nevertheless treated a double free in Regexp compilation as a vulnerability after a comprehensive assessment
not sure what sort of comprehensive assessment is still called for at this point?
We've git that under control with regulation and the help of organizations, some of which still exist today.
This "era of constant exploits/patches" is not normal and can't go on. Currently it is an ongoing cat and mouse game, but it doesn't have to be.
Security is always reactive by its very nature. You can do proactive things, but you will always be reactive to novel attacks.
Expect everything to have various levels of rolling upgrades, and soon. That's a lot of work.
When it comes to software it hasn't always been the case. Before the Internet software was considered something stable and even now there are areas, like functional safety critical software, where this is true to some degree.
It's not that it can't be done.
It used to be expensive to a certain extent it will stay expensive, but modern languages, tooling and formal methods will hopefully mitigate that enough.
The future is near constant burning of tokens in your software factory to do this automatically. The labs create the increase in demand for their own product.
The way LLMs are developed now is but one possible future we could have experienced. Imagine the future where the LLMs were built by a few governments with black budgets and hidden until a war or some other large event broke out. Then suddenly most of the computers and digital infrastructure shits itself as everything goes wonky.
At least this way we get to see that a shared global infrastructure built on the idea that security is someone else's problem was a really bad idea and we might have enough time to fix it or pull enough plugs we don't go back to the dark age.
Alongside the current scanning and patching of software.
I don't think that lasts, I've seen at least 5 viable replacements now. All internal projects at major US companies.
This might feel like a "Yeah obviously" moment, but the fact is that very few Rails apps are running the latest ruby release the moment it comes out.
I mean - this incident brings it into sharper focus - but it's always been thus.
As a Go developer I often tell people "Be wary of vendoring, you are creating a cache of dependencies that need to be managed and maintained"
Or people that "pin" dependencies, they face the same difficulty.
Why not just use "official" (in this case) images - sometimes they lag too - hell, most Linux distributions lag so much with Go compiler packages that, again, I routinely tell people to avoid them in favour of watching the Go official page
Who else is tired of C libraries since this week?
And now we are seeing that was all untrue.
How about being constructive instead - old bug in old code is no surprise, no matter the language/technology.
https://www.darpa.mil/research/programs/translating-all-c-to...
Getting rid of C isn't the holy grail it might appear to be (Ubuntu has been working hard at rewriting sudo and coreutils in Rust and they have had plenty of bugs in the process), but I do hope the C projects will preemptively start fuzzing/running fil-C/invent new safety mechanisms for old code.
Rust will not protect you against exploding regexes, it will not protect you against integer over/underflows (unless you specifically call safe wrapping methods), but it will protect you against double frees if you use it sensibly (nothing prevents you from breaking the borrow checker and messing up memory if you write an unsafe {} block).
But yes, memory-save languages like Rust are a massive help protecting against remote code execution. (unless said memory-safe language introduces its own category of RCE vulnerabilities, e.g. deserialization in Java)
shouldn't they tag a new version instead of re-releasing a fixed 19.3.6, if earlier 19.3.6 are vulnerable?
Just get the source code of Forgejo, audit it thoroughly and run an instance of that.
I like Forgejo btw, but these days everything is vulnerable. Switching tools isn’t the final answer
In this situation, wouldn’t the more responsible behaviour of gitlab be to blacklist Ruby runtimes that enable the CVE?
I do think CVE-2026-89078 - the double free one - may only affect ruby 3.3.11 and above. It relies on a bug introduced in ruby 3.4/4 via https://github.com/ruby/ruby/commit/8cd295ecf0 and backported (maybe accidentally?) in 3.3.11/12 in https://github.com/ruby/ruby/commit/e02c892ade. So that one is getting patched on all rubies it would have affected.
Linux has the very same disturbing not invented here syndrome, rewriting good features by their own, making them worse.
Surely we're approaching a point where an LLM discovers a critical, unfixable vulnerability in every device running any version of $OS, then irresponsibly discloses it, triggering mass collective trauma among admins worldwide, significant downtime in critical infrastructure and scores of deaths.
We joked in our last internal meeting that this is the end of public software... I have a feeling we may be right, at least in the very short term.
https://news.ycombinator.com/item?id=50189092
Thankfully a lot of the low level stuff is just not that exposed to the network so even if you find a bug in some firmware, you still have to find a way to access it over the network.