At this point that’s not unexpected. Given the proliferation of tools generally, if not freely, available, it would be naive to think that the published PoC was the only one in existence.

As a result, the curl project has just changed their CVE policy to publish all with a severity less than HIGH, starting 26 October.

See:

https://github.com/curl/curl/discussions/23458

The server has been hugged to death … I wonder why 😁

It always was.

I moved from VMware Fusion after a hardware failure and migrated to Proxmox. I use VMs to isolate client information and to test and use different OS versions and distros.

I have a VM that runs Docker and connect to it via SSH for running applications I don’t want to install on my main workstation.

My main workstation is also a VM.

I came from bare metal Linux for years and found that I spent too much time recompiling kernels to make it work with my hardware.

Now that everything is a VM, changing underlying hardware, and even hypervisor is trivial.

If VMware hadn’t been bought by Broadcom and allowed me to virtualise x86 on Apple Silicon, I’d still be running VMware Fusion. UTM by the way is completely unreliable … I really tried.

As it is, I’m running Proxmox on AMD.

in summary, Proxmox allows me to build what I need when I need it virtually and it’s running on a standard Debian host.

Edit: #$#__# autocorrect (unreadable -> unreliable)

PHP is written in text files which can be managed using any number of version control systems. Pick your poison.

https://en.wikipedia.org/wiki/List_of_version-control_software

As someone else said, you can document your patches with a Docker file, but you’ll still need to track your changes.

Ultimately, what you’re essentially doing is maintaining a local fork of the code and keeping it synchronised with upstream.

We told you so …

https://www.techdirt.com/2026/02/25/hackers-expose-the-massive-surveillance-stack-hiding-inside-your-age-verification-check/

What are you trying to block?

I’m asking because based on the traffic I saw on a live retail website had malicious access attempts coming from significantly more addresses than you seem to be indicating and more often than not the address was different for each request.

That said, Meta DDoS bots were outright blocked because they hit the same URL thousands of times for no apparent reason.

Note that this was during the first half of this year, the landscape has undoubtedly worsened since then.

For how much data?

What are you using to get it there and back?

I’m asking because I had been using this for years and only after a catastrophic event, discovered that my “helpful” backup software had turned on versioning which AWS, also helpfully, doesn’t explicitly tell you when you attempt to delete old backup sets, showing no objects but charging you anyway.

I did recover all my data after a catastrophic event, but it was an expensive experience to put it mildly.

Unless I misunderstood your original question, you don’t trust the Docker signing process, so getting a cryptographically signed image from there seems to fail your criteria.

You could build a Debian base image in-house from scratch if you wanted to and limit your tools to Debian only signed packages.

Ultimately you need to trust something.

For example, how do you trust the microcode updates to your CPU?

I tend to build my own containers for anything that matters, but that might not be a viable option.

Microsoft continues to be the best advocate for migrating to Linux.

26 years ago.

https://lists.debian.org/debian-sparc/2000/09/msg00038.html

“That’s not a flaw … that’s a feature!”

Around the turn of the century I built a “middleware” layer for an Oracle to MySQL bridge which enabled legacy systems to keep running while modern and open source tooling could be built.

The biggest challenge was moving from columnar to row based data … that and compiling Solaris tools using Oracle libraries.

Oracle is like sendmail, user friendly, just very picky whom it makes friends with.

I’m getting flashbacks to debugging an Oracle connection file.

… and replace them with a whole new class of bugs.

Source: I’ve been writing software since 1982.

And if you think that migration to Rust will magically remove all bugs and vulnerabilities, you haven’t been writing software for very long.

Seems like a valid use of the technology…

The website has a heading proclaiming that apparently it’s a bad year for the Linux kernel.

I have to ask … compared to what … the Windows kernel?

Wow … so I can finally have root on my own phone?