@filippo@abyssdomain.expert
@filippo@abyssdomain.expert avatar

filippo

@filippo@abyssdomain.expert

@FiloSottile elsewhere / Cryptogopher / Go crypto maintainer / Professional Open Source maintainer / RC F'13, F2'17

https://mkcert.dev / https://age-encryption.org / https://filippo.io/newsletter

🕳️ “Gaze not into the abyss, lest you become recognized as an abyss domain expert, and they expect you keep gazing into the damn thing.” —@nickm

This profile is from a federated server and may be incomplete. Browse more on the original instance.

osxreverser, to random
@osxreverser@mastodon.social avatar

So, Go's infrastructure appears to be running an arbitrary Internet crawler based on git and other VCS and people don't appear to be excited with the RCE potential? Guess I'm getting too old... :P

filippo,
@filippo@abyssdomain.expert avatar

@osxreverser we assumed it was fully RCEable from the start :) it’s properly sandboxed.

filippo, to random
@filippo@abyssdomain.expert avatar

You might know I am a recurse.com alum.

I love that community. Full of smart, kind, passionate folks.

For the first time this year I was on the hiring side of their recruiting services, and it still beat my expectations.

The RC-sourced applicant pool was amazing.

I moved forward 57% of RC applications (and even the ones I rejected were great, just not good fits), versus 18% of non-RC appls.

Way more diverse pool, too.

tl;dr: if you need to hire nice, smart folks quick, reach out to RC.

filippo,
@filippo@abyssdomain.expert avatar

It looks like there aren't a lot of companies hiring for security/cryptography roles from Recurse right now, which is a hell of an opportunity. Every batch has folks doing Cryptopals and CTFs!

You know how in security hiring what matters is finding the sharp curious folks, especially at the junior level? Exactly.

filippo,
@filippo@abyssdomain.expert avatar

Always happy to answer questions about RC if anyone has them. DMs are open.

To be clear I am not affiliated with the faculty, I am a (returning) alum. I always recommended attending RC, now I engaged them on the hiring side, too, and can strongly recommend hiring from them as well.

filippo,
@filippo@abyssdomain.expert avatar

@ljrk RC is fully hybrid now! If remote collaboration works for you (which will depend more on yourself than RC) you should absolutely consider it. The virtual RC tooling is top notch, and the faculty and community really care about including remote folks.

It is absolutely not MENSA-style "smart". I understand the vibes, but they do a lot of work to fight impostor syndrome and to not define "smart" in the mainstream """measurable""" stereotypical way.

osxreverser, to random
@osxreverser@mastodon.social avatar

index.golang.org returns some results with uppercase names and if you use that to query sum.golang.org you will get bad request errors. I hate computers :P How can everything be so dumb? So much wasted potential 😠

filippo,
@filippo@abyssdomain.expert avatar

@osxreverser uppercase letters are encoded as !x to match the filesystem path, which can’t be case sensitive. It’s documented :)

filippo,
@filippo@abyssdomain.expert avatar

@osxreverser I’m from mobile, but /ref/mod and the sumdb design documents might be it.

In general, use “go mod download -json name@version” or emulate https://go-review.googlesource.com/c/exp/+/321030

Plz use cached-only, see https://proxy.golang.org

filippo, to random
@filippo@abyssdomain.expert avatar

Thinking of doing a livecoding session to integrate the new Kyber/ML-KEM post-quantum key exchange in Go’s crypto/tls. https://github.com/golang/go/issues/67061

Anyone interested? Would probably start in about an hour, say 1400 UTC. https://twitch.tv/filosottile

filippo,
@filippo@abyssdomain.expert avatar

Live now! Integrating the new Kyber post-quantum key exchange in Go’s crypto/tls package.

https://twitch.tv/filosottile

filippo, to random
@filippo@abyssdomain.expert avatar

Strong agree that sudo is dogma, and logging in as root is just fine, actually.

I think @fanf is even more right about this than he claims.

For single-user workstations, who cares about administrative access. The only real security boundary is the TPM/SEP. really(8) without any further authentication would be just fine.

The flip side is that I don't actually care about sudo's complexity or security, because it's not protecting a security boundary I care about.

https://dotat.at/@/2024-05-02-sudo.html

filippo,
@filippo@abyssdomain.expert avatar

@agowa338 @fanf Fair, sandboxing is also a real security boundary, but it needs to be a carefully constructed one, not just "make a non-root user that can still see all the filesystem and execute all the SUID binaries" which is what sudo protects.

filippo,
@filippo@abyssdomain.expert avatar

@agowa338 @fanf Disagree, there are robust boundaries and symbolic ones.

TPM/SEP with PIN and user presence checks are very robust. I could give you root on my machine and you could not extract my SSH key.

Restricted sandboxes with namespaces are reasonably robust. We run untrusted Javascript every day. Bypasses will get you large bounties.

Unprivileged but unrestricted users on most machines are basically as useful to an attacker as root.

filippo,
@filippo@abyssdomain.expert avatar

@agowa338 @fanf Most microarchitectural vulns compromised SGX-like systems, not discrete TPMs, and I am not aware of any SEP extraction… ever? Same for YubiKeys.

I also very much mean keys that live in the TPM/SEP, not in memory wrapping. Check out age-plugin-yubikey and Secretive.

filippo,
@filippo@abyssdomain.expert avatar

@agowa338 @fanf "Need physical access" is a hell of a security boundary, yes.

filippo,
@filippo@abyssdomain.expert avatar

@cks @fanf Sure but really(8) is enough for that, without reauth.

filippo,
@filippo@abyssdomain.expert avatar

@cks @fanf I mean, replace really(8) with a bash script that does a TTY confirm read and then calls really(8). It’s still not justifying the existence of sudo as a complex ACL based tool with reauth capabilities.

filippo, to random
@filippo@abyssdomain.expert avatar

In Go 1.22 we fixed a common security footgun by replacing the math/rand[/v2] default generator with a new secure one based on ChaCha8.

Details on the Go blog by @rsc and me (and full reusable specification in C2SP).

https://go.dev/blog/chacha8rand

atoponce, to linux
@atoponce@fosstodon.org avatar

Just sent in my first patch to the #Linux kernel.

This changes the kernel CSPRNG from ChaCha20 to ChaCha8 providing ~2x performance improvement without sacrificing security.

#crypto #cryptography

https://lore.kernel.org/lkml/20240429134942.2873253-1-aaron.toponce@gmail.com/T/#u

filippo,
@filippo@abyssdomain.expert avatar

@atoponce lol

IIRC BoringSSL had a userspace CSPRNG because the kernel one was too slow for TLS record IVs. Check if that’s still the case?

(Ted is wrong, in TLS 1.2 you need random at every record, not just when doing asymmetric crypto. TLS 1.3 fixed that.)

hanno, to random
@hanno@mastodon.social avatar

Does Python really have no DNS functionality built in at all beyond resolving IPs? I have a use case where I need to get a TXT record, and everything I can find recommends dnspython. If possible, I'd like to avoid adding a dependency.

filippo,
@filippo@abyssdomain.expert avatar

@hanno name to IP resolution is provided by the OS, but other DNS lookups are less abstracted tasks so it’s common for them not to be provided by stdlibs, for better or worse

benjojo, to random
@benjojo@benjojo.co.uk avatar

TIL you can open MP4 (container) files in wireshark and it sorta works... I guess it's all "packets" at the end of the day

filippo,
@filippo@abyssdomain.expert avatar

@benjojo File formats are protocols!

filippo, to random
@filippo@abyssdomain.expert avatar
filippo, to random
@filippo@abyssdomain.expert avatar

We're getting hedged signing and domain separation in all NIST post-quantum signature algorithms!!!

https://groups.google.com/a/list.nist.gov/g/pqc-forum/c/JKMh0D0pa30/m/vbflXolxAQAJ

Also, derandomized testable APIs for all of the NIST PQC algorithms.

https://groups.google.com/a/list.nist.gov/g/pqc-forum/c/Mf2kemwwreY/m/KArjoIhxAQAJ

It's basically Christmas 🎁 🤩

filippo, to random
@filippo@abyssdomain.expert avatar

A PSA since there's some confusion on this...

There is no vulnerability in Gorilla Sessions.

The vulnerability is in Palo Alto's internal SessDiskStore, which looks similar to FilesystemStore. Early analysis came to the mistaken conclusion that the vulnerable path was in FilesystemStore, but it's not. FilesystemStore authenticates the Session.ID with securecookie, SessDiskStore does not.

filippo,
@filippo@abyssdomain.expert avatar

Hypothetically, if an application went out of their way to misuse FilesystemStore by not using its New API and stuffing attacker-controlled data in Session.ID (which is documented as not being safe), they could hit this.

That's not what happened to Palo Alto. They wrote their own Store that takes the session ID from a cookie in New without authentication.

mcc, to random
@mcc@mastodon.social avatar

I have literally implemented SRP at both the client and server side but I am still unable to figure out, if I were to purchase or set up a "Passkey", what exactly I would have, or how it would work, or which computers, web browsers or web sites I should expect it to work with

filippo,
@filippo@abyssdomain.expert avatar

@irenes @mcc @glyph @djc right, I guess I should say passkeys are primarily meant to replace passwords, and they happen to also be replacing U2F because the APIs and UI flows are unifying, but they don't start out as a U2F replacement, so it makes sense they wouldn't feel as the right way to replace U2F in isolation.

  • All
  • Subscribed
  • Moderated
  • Favorites
  • megavids
  • modclub
  • DreamBathrooms
  • mdbf
  • Durango
  • ngwrru68w68
  • magazineikmin
  • thenastyranch
  • InstantRegret
  • Youngstown
  • slotface
  • everett
  • kavyap
  • cisconetworking
  • JUstTest
  • ethstaker
  • tacticalgear
  • GTA5RPClips
  • osvaldo12
  • khanakhh
  • rosin
  • Leos
  • normalnudes
  • anitta
  • cubers
  • tester
  • provamag3
  • lostlight
  • All magazines