> RustDesk still does not support encrypted connections when self hosting
Not true at all. RustDesk connections are fully encrypted even when using a self-hosted server. What is not supported are encrypted direct endpoint-to-endpoint connections not going through a server at all, because that's not a use case RustDesk is designed for. The feature is intended for testing, which is why it's disabled altogether by default.
If your use case is focused on direct connections without a server, RustDesk is the wrong tool for the job. It's an open-source self-hostable analog to Teamviewer, the server is inherent to the design.
SubiculumCode 20 hours ago [-]
I think you bring up a fine point. Most applications have encryption built in and transparent to the user. While technical IT people in charge of infrastructure will not make this assumption, it is not unreasonable, at the same time, for small business / individual users, to not understand or know of this limitation.
VNC does not, for example, and you need to set up port forwarding through ssh.
X2Go does....but I don't trust it terribly. So, RustDesk..what do you need to wrap it in? ssh?
I really want a better solution that VNC/X2Go..especially since I use hardware accelerated apps that need to use VGL to operate correctly (but underwhich operation is quite buggy).
NewJazz 10 hours ago [-]
You could wrap in a wireguard tunnel.
dj0k3r 20 hours ago [-]
Tailscale, or any encrypted mesh overlay is perfect for this. Infact I prefer it that way. Rustdesk can do what it does best at its core.
preisschild 19 hours ago [-]
I disagree, modern software should make encrypted connections over something like HTTP3 or QUIC directly so true secure end to end connectivity works. This would make VPN software such as tailscale obsolete.
mnahkies 16 hours ago [-]
One of my primary use cases for tailscale/VPN is that I can happily run stuff (grafana, gitea, etc) and not have to be panicked about monitoring for CVEs - I serve it all over HTTPS but I don't want to put it on the public internet if I don't need to.
preisschild 43 minutes ago [-]
I just have envoy proxy with the oauth + jwt filter in front of those services. Envoy does the oidc flow with pocket-id so I can use passkeys for authN. Envoy validates the resulting token and does authorization via ACL. Envoy then sends an authorization bearer jwt with the oidc id_token jwt to the backend (for example grafana). Grafana parses and validates the jwt and sets claims as userinfo (username, groups, email).
I think such setups are at least as secure as having tailscale in front of it and they are web standards conform. I dont need a client app like tailscale, I can just use my normal browser and internet conn.
I always make sure envoy/all other apps are on the latest security patcb anyways.
ZoomZoomZoom 4 hours ago [-]
Secure and censorship-resistant connectivity is very hard. We should compartmentalise and not expect everything to reimplement and maintain their own version of it and focus on specialised solutions that actually work.
What all user-facing software should have is a minimal-overhead connection option to improve performance inside user's tunnel of choice.
Pure QUIC gets blocked easily, SSH requires wrapping, even Tailscale mimicry is basic and they still ignore simple protocol improvements available.
elevation 19 hours ago [-]
I'm also in favor of adding encrypted connections to RustDesk, not to replace tailscale, but as a part of this complete breakfast. Tailscale provides mutually authenticated, authorized L3 access. TLS can be configured to provide mutually authenticated L4 access. With a little OIDC/webauthn setup, both L3/L4 support device attestation, meaning there's no way to connect without e.g. a yubikey (or perhaps an enrolled TPM.)
stavros 5 hours ago [-]
Doesn't Iroh do that? It would be great to go back to the peer to peer days, but with security.
thatfunkymunki 19 hours ago [-]
agreed, zero-trust solution with proper endpoint security and PKI is superior to transport encryption and insecure protocols
jcelerier 17 hours ago [-]
so without tailscale or any other intermediary turn service how does my computer behind a NAT connect to another computer behind another NAT
preisschild 16 hours ago [-]
Ideally using ipv6 without NAT, but yeah hole punching or quic address discovery like iroh does works too
LoganDark 9 hours ago [-]
and how do you discover the ipv6 address? memorize it? many ISPs don't listen to informational documents listing all the problems with random dynamic prefixes.
jcgl 6 hours ago [-]
Dynamic DNS is fine for many use-cases.
LoganDark 4 hours ago [-]
I suppose, but then everybody else knows your IP too, which nullifies one of the benefits of IPv6.
theultdev 17 hours ago [-]
by using something like iroh.
basically an embedded tailscale.
teekert 10 hours ago [-]
Tailscale also protects against login attempts.
marshray 18 hours ago [-]
So I'm supposed to set up PKI before I can open a remote console connection?
Please just make it work seamlessly with my existing SSH credentials. Like SFTP.
kaoD 7 hours ago [-]
You don't need PKI.
Nothing prevents RustDesk from implementing TLS TOFU (see e.g. Gemini Protocol), which offers the same security guarantees as SSH TOFU.
preisschild 17 hours ago [-]
You could just make use of public oidc/oauth2 providers like Google/Github/Microsoft/Cloudflare Generic OIDC or heck, even your bluesky account via ATProto oauth
But yeah, you could also make use of your ed25519 ssh public key as client certificate and accept based on fingerprint like ssh
javier2 6 hours ago [-]
managing the certificates safely is too much of a hassle. especially if i dont want to accessible on public internet.
tomjakubowski 17 hours ago [-]
Would be nice if they did support that, but on a LAN you can always encrypt at layer 3 with WireGuard.
trueno 9 hours ago [-]
their reply in that issue:
"We don’t plan to implement this directly, as it is designed for test initially, we do not recommend using RustDesk with a public IP alone, that's also why direct ip access is turned off by default.
For security and privacy, we strongly advise using it in combination with a VPN—the VPN tunnel already provides end-to-end encryption, so additional encryption within RustDesk is unnecessary in that setup.
That said, we welcome community contributions!
Many users express interest or share feedback—but very few take the next step and submit a pull request.
If you’re able to help, your PR would be greatly appreciated! "
LoganDark 9 hours ago [-]
> That said, we welcome community contributions! Many users express interest or share feedback—but very few take the next step and submit a pull request. If you’re able to help, your PR would be greatly appreciated!
hate this. "we don't want this on principle... but we would accept it if you did it yourself!" just say you don't feel like it! don't act like it's a bad thing except for when somebody else does it for you, that's just dishonest.
embedding-shape 3 hours ago [-]
> hate this. "we don't want this on principle... but we would accept it if you did it yourself!" just say you don't feel like it! don't act like it's a bad thing except for when somebody else does it for you, that's just dishonest.
What is there to hate here, I don't understand?
They first provide a concise overview of the current situation. Then how they personally feel about it. Finally they leave it open and end it with "If you want it badly enough, we'll accept outside contributions for it".
They're not saying that they're against the feature in principle, they're saying that they themselves don't plan to spend time implementing it, but if users really want it and contribute the feature itself, they'll be happy to maintain it once merged.
This seems like the ideal solution? What would be better here?
trueno 8 hours ago [-]
yeah it's definitely confusing messaging i can't imagine being a community contributor there and saying "ok im gonna help implement this thing that they just described as not a good fit for their thing". i think being under resourced or whatever or just not having anyone on the team that would be able to pull this off is a fine thing to be forward about, but it gets confusing because they kinda frame it like it's the wrong thing to do then leave the door open for someone else to do it. if they didn't beat around the bush perhaps they'd have encryption on this already.
LoganDark 8 hours ago [-]
I prefer when projects say "we don't have the resources for this but we agree it would be nice" because then I actually feel like it would be worth the PR. but this project makes me feel like they don't want the change which means I don't want to put in the work
8 hours ago [-]
innocent_name 17 hours ago [-]
>does not support encrypted connections when self hosting
Factually incorrect. If you self host a relay/coordination server - encryption works as documented.
guiambros 12 hours ago [-]
Only if you use a relay server. Direct connections in a local network are not encrypted, and RuskDesk doesn't plan on implementing it [1].
Your statement is not entirely accurate. This only applies when using Direct IP Access on local networks, which is off by default. Their justification and invitation to PRs is the final comment [1] on the issue you linked. Why are you leaving this information out of your comment?
It may only apply to Direct IP Access, but isn't that what many would expect (not trusting middleman to be secure?)
innocent_name 17 hours ago [-]
Have you tried reading the docs?
>RustDesk is a single open-source application (AGPL) with its own architecture. Clients connect outward to an ID/rendezvous server, which brokers a peer-to-peer or relayed session. Per the RustDesk documentation, traffic is end-to-end encrypted (built on NaCl)
SubiculumCode 17 hours ago [-]
Not yet...good to chastise me on this, but in my defense, I am at work, and am just trying to get a sense of this, as I would like to move away from VNC.
preisschild 19 hours ago [-]
I still think its an essential feature for self-hosting, but they seem to be open to PRs so what is the problem?
nextaccountic 18 hours ago [-]
[dead]
OsrsNeedsf2P 15 hours ago [-]
Rustdesk is amazing. I was using it just 2 days ago and ran into this hiccup, so it's a pleasure to see it resolved
throwaway27448 21 hours ago [-]
What is rustdesk and how is it distinct from vnc?
Edit: i appreciate the explanations; thank you.
pizza234 20 hours ago [-]
VNC and RustDesk are both remote desktop solutions, however, Rustdesk is considerably more performant than the VNC family, because the latter primarily sends framebuffer updates, while the former can use modern video codecs and temporal compression to encode screen changes much more efficiently.
shock 19 hours ago [-]
> Rustdesk is considerably more performant than the VNC family
It consumes, on the client, 800%-1200% CPU AND 8%-10% GPU decode on my NVIDIA card. I opened a bug and they transformed it into a discussion, without any response.
embedding-shape 3 hours ago [-]
> they transformed it into a discussion, without any response
How dare they.
diego_moita 18 hours ago [-]
I just have one word for you: TightVNC.
That's what made me uninstall RustDesk.
pizza234 17 hours ago [-]
I have three words for you: read the docs.
If you're experiencing a bug that pegs you're CPU, fair enough. But RustDesk is video-stream based (in many formats), while TightVNC is image(-region) based. This is actually similar to why you watch movies in, say, H.264 rather than Motion JPEG¹.
¹=similar, not same.
diego_moita 8 minutes ago [-]
> I have three words for you: read the docs.
No, thanks.
As they say: "the proof of the pudding is in the eating". So I used both applications and TightVNC is just a lot better, particularly in constrained remote VMs.
> TigerVNC was originally based on the (never-released) VNC 4 branch of TightVNC.
It seems to be popular, that’s why I’m asking.
diego_moita 11 minutes ago [-]
Thanks, I never tried it. Will try.
Funny thing is that the document where they'd explain their reasoning to fork TightVNC returns 404.
vablings 21 hours ago [-]
Well first there was TeamViewer which was VNC with more bells and whistles then it became enshittified. Then anydesk came along and ate up teamviewer then that became enshittified, Now we have rustdesk which seems to hopefully be a bit more immune to being enshittified.
If you already use VNC this is not something for you
mschild 19 hours ago [-]
RustDesk is also open source including the server so you can fully self-host.
pyramidless_cle 16 hours ago [-]
That's not really true for quite some time now. Both client and the free server depend on a separate submodule that has no LICENSE file whatsoever. Because of that RustDesk is stuck in a limbo somewhere between open source, source-available and closed source. Also, the whole management of the project seems questionable. Also for some features a dll blob which is not open source is required with no source disclosed at all.
I like that their server is just a single Go binary, but then they go and ruin that with a Node.js web UI and.. PostgreSQL?! Seems like complete overkill. I wonder what gripping limitations they faced with SQLite that they decided it needed PostgreSQL.
opello 10 hours ago [-]
It looks like PostgreSQL is optional and SQLite is the default.
SQLite is still the default backend. Postgres is optional.
pyramidless_cle 3 hours ago [-]
Seems like that is just a server implementation that requires you to use stock RustDesk clients. So you can't really have a fully opensource RustDesk-like system without requiring a substantial chunk of unlicensed source code as well as a closed source dll blob(s).
I just glanced over the repo, deployment and comments around it and it seems like unconstrained AI slop or at least the project seems to be led by a person which can't use AI effectively for the purpose of software engineering.
topspin 19 hours ago [-]
I've noticed that RealVNC has hidden their formerly free VNC desktop client, now called RealVNC Classic Viewer, behind an enterprise subscription login, and replaced it with a different cloud first thing, RealVNC Connect Viewer, that requires a login to at least obtain, and is anemic in terms of features.
Not that RealVNC is VNC. However, enshittification has clearly taken another scalp.
rcxdude 20 hours ago [-]
VNC is one way of doing remote GUI access (well, more a family of different protocols and products with different capabilities and tradeoffs). RustDesk is just another product that does the same thing but doesn't have a specific connection to VNC as a protocol so e.g. doesn't have to handle legacy authentication modes and stream formats. In my experience most remote access solutions beat VNC for performance, for example.
aborsy 17 hours ago [-]
I have used Remmina over SSH and Tailscale (and generally don’t trust remote desktop tools), which is mentioned in various places by my operating system provider.
How does RustDesk compare with that?
fsh 17 hours ago [-]
It's not VNC-based, so the performance is much better. It also supports multiple monitors.
aborsy 16 hours ago [-]
Thanks!
GNOME Remote Login runs an RDP server, and RDP indeed isn’t fast enough, although is TLS-encrypted.
SubiculumCode 11 hours ago [-]
Isnt the issue with the latter is tha tyou need a local session first before you can connect?
boardwaalk 19 hours ago [-]
Is there a reason to use this over something like Sunshine/Moonlight?
himata4113 18 hours ago [-]
rustdesk just works, moonlight and sunshine is janky as hell outside the local network due to the fact it uses like 5 different streams.
SubiculumCode 18 hours ago [-]
I had a heck of a time ever getting that to work. Also, can that support multiple concurrent users?
pineapplepizza6 15 hours ago [-]
[dead]
manav 21 hours ago [-]
Does my screen have to be on?
NoboruWataya 17 hours ago [-]
I have a Raspberry Pi hooked up to my TV and whenever I need to control it from my laptop (getting and interacting with the same screen on my laptop that I see on my TV) I have been using VNC. It is quite slow. Would this work better for that use case or is it really intended for something different? (And if the latter are there better solutions than VNC for what I want to do?)
c-hendricks 17 hours ago [-]
Yes this will give you similar results.
There's VNC and VNC-likes (rdp, steam link, rustdesk, etc) which all do similar: stream a screen, allow remote keyboard/mouse, maybe redirect sound, maybe use video encoders.
There's also things like deskflow, which allow you to use one keyboard/mouse and control with multiple machines.
NoboruWataya 16 hours ago [-]
Thanks! I will check it out then.
Kavelach 17 hours ago [-]
If you can look at your TV instead of your laptop, I highly recommend using KDE Connect
andai 21 hours ago [-]
Menu css looks a bit dodgy with the notification at the top:
Does it framebuffer grab the current session and inject input events?
jchw 18 hours ago [-]
Usually the way to be able to even get the login screen is to use DRM/KMS based capture and use uinput for device emulation. It doesn't strictly need root but it does need privileges a user wouldn't have, like access to the DRM card nodes and uinput device. Sunshine does it this way.
It entirely sidesteps the problem of X vs Wayland, although this does mean the unattended physical seat can be physically taken over while you are remote.
codedokode 16 hours ago [-]
As I remember, generic DRM interface allows only writing to a framebuffer and not reading from someone's else framebuffer. Do they use some vendor-specific APIs?
jchw 4 hours ago [-]
Not that I'm aware of. I'm on my phone right now but any KMS driver that supports KMS and dma-bufs should work.
The basic idea is to enumerate the CRTCs to get the relevant framebuffer plane resources and then use dma-buf exports to pass them to an encoder. Reality can be a bit more complex due to the existence of overlay planes but that's the idea.
Yea I need to know this before I even consider something like this. I'm guessing it's a daemon that runs as root in the background.
SXX 16 hours ago [-]
Btw does RustDesk support microphone input passthrough from client to the host yet?
So far it's only feature I found missing compared to proprietary solutions.
j_aime 14 hours ago [-]
Unfortunately, no, it's not working. I use Rustdesk all the time. I've found a way, but it's really stupid. I have a GL.iNet Comet to manage my headless machine. That unit supports microphone passthrough. I send my mac's mic through that to ubuntu. It's an awful solution hah, but it works. I'm able to use dictation through that on Ubuntu. I didn't buy the comet specifically for mic use. I found out after the fact that it was able to do it.
It's open-source, so just build it yourself with the tiny change. Something this trivial could be done with a 30 second prompt at this point, so there's not even an excuse of "too much effort".
I will note that the XKCD password scheme being proposed there is, in fact, completely insecure. A modern consumer GPU can crack "four random English words" in a day. You can argue that it's the user's choice to be allowed to use insecure passwords, but arguing that that scheme is actually secure is just wrong.
bilkow 19 hours ago [-]
> A modern consumer GPU can crack "four random English words" in a day.
Let's run the math:
EFF's long wordlist[0] (the one used by Bitwarden's passphrase generator) has 7776 words, which is about 13 bits of entropy per word (log2(7776) = 12.92). A 4-word passphrase then has 51.7 bits of entropy, meaning there are 2^51.7 possible options for the passphrase, which is 3.66E15 possible options.
For reference, a password with all completely random characters and symbols, for each character you have 70 possibilities, or about 6 bits of entropy. An 8-character completely random password (no words or the usual patterns) then has about 49 bits of entropy, which is less than a 4 word passphrase.
About the time it takes to crack it, assuming you can test an average of 1 password per μs (you probably can't, as passwords are usually stored using key derivation functions[1][2] with work factors tweaked for current hardware) it would take ~116 years to crack it. I usually see 5-passphrase recommended nowadays, which would multiply the required effort by 7776.
Sure, a 16-character random password has a lot more entropy, but it's also a lot harder to interact with (reading, comparing, typing) if you end up needing to, and it's still easier to crack than an 8-word passphrase.
> assuming you can test an average of 1 password per μs (you probably can't, as passwords are usually stored using key derivation functions
You are not (usually) in control of what services use to encrypt your password, and most software in the world uses a basic hash like MD5 or SHA-256 rather than a key derivation function, meaning you are wildly, wildly underestimating the number of guesses per second here. To suggest that passwords are "usually" stored like that is just wildly incorrect. Only well-educated, high-quality engineers who care about security, a vanishingly tiny portion of the industry, even think about things like this. I still run into government services that e-mail me my password in plaintext.
---
Sorry, but edit attaching a reply to another comment here because HN rate-limits me:
> This is just completely false.
My mid-tier laptop GPU can crack the usual wordlist in under 12 hours against MD5 (46 billion guesses per second), or about 3 days against SHA-256 (6.5 billion guesses per second). A desktop 4090 would likely crack SHA-256 in under a day (on average).
I make claims with confidence because I am speaking about facts. Why are you making claims with confidence when you have no idea what you're talking about? The quality of discourse on this site is so, so low.
wavemode 17 hours ago [-]
What you said:
> A modern consumer GPU can crack "four random English words" in a day
What you're saying now:
> My mid-tier laptop GPU can crack the usual wordlist in under 12 hours against MD5 (46 billion guesses per second)
So your response to the plain mathematical fact that, no, your consumer GPU cannot crack "four random English words" in a day, is that your consumer GPU can crack four words chosen from a list of a few thousand, in a day. Followed by personal attacks. Okay.
BenjiWiebe 15 hours ago [-]
Well, there could be quite a range of performance between a mid-tier laptop GPU (his consumer GPU), the average consumer GPU, and the average (gamer's) consumer GPU.
17 hours ago [-]
drdexebtjl 17 hours ago [-]
Great, then use 5 words instead.
Trying to pick a strong password assuming the service you’re signing up for doesn’t care about security is pointless. You’ll be pwned regardless of what you do.
mr_mitm 18 hours ago [-]
> most software in the world uses a basic hash like MD5 or SHA-256 rather than a key derivation function
For passwords? Where do you get that information?
applfanboysbgon 17 hours ago [-]
Experience? Looking at codebases? The one in question does, in fact, use SHA-256. HN is a bit of a bubble because there's a large demographic of FAANG-tier employees here who work at big tech or unicorn startups, but most software in the world is not made by FAANG or unicorn startups, and the standards are way, way lower than you might think they are if you're used to living in this bubble. For software development practices in general, not just password storage. I think it's taken for granted that every single person on HN uses VCS, but even that is a luxury in many real-world working environments outside of SV.
Even if only 10% of the services you use use MD5/SHA-256 (although I certainly would expect it to be >50% if we could do a large-scale study), why accept your password being easily compromised 10% of the time?
pbhjpbhj 3 hours ago [-]
>why accept your password being easily compromised 10% of the time?
Those services using md5 aren't doing important things, I hope, so the password for that service hardly matters.
If it does matter, then user long passwords/phrases and rotate every time (or find a better supplier).
tredre3 21 hours ago [-]
> A modern consumer GPU can crack "four random English words" in a day. [...] but arguing that that scheme is actually secure is just wrong.
Let me do just that!
This is a networked service. You send your password (or a hashed form) to it, and it validates it. You don't have the local hash to bruteforce it offline.
Even if we only consider the top 10k english words, it's 10000^4. It's going to take years to bruteforce this over a network because you'll go through so many rate-limits, cooldown periods, and outright bans that it's questionable whether it's even possible.
applfanboysbgon 20 hours ago [-]
Virtually any password other than "password123" is fine if you're rate limited to a few guesses per day by a networked service. Passwords should be secure against the inevitable data breach when the service you're using loses their hashed password database, which happens on a routine basis.
strbean 19 hours ago [-]
Seems like the only real issue here is that rustdesk uses SHA256 instead of argon2.
tialaramex 20 hours ago [-]
> This is a networked service. You send your password (or a hashed form) to it
Hopefully neither. But given everybody involved in VNC seems to be the same batch of clueless morons who built all those PHP web forums twenty years ago with MD5 as the password hash, who knows what they cobbled together. Maybe an expert can chime in about what actually happens here?
Yes, for a sensible scheme this can't work.
minitech 17 hours ago [-]
I’m not an expert in RustDesk, but I did take a quick look at the code, and it’s filled with massive red flags like [1], where it appears the general mechanism is SHA256(SHA256(password++salt)++challenge). I can’t even tell if the whole thing is vulnerable to trivial MITM ([2] makes it sound like it might be; while investigating, I ran into the concerning comment [3], “fall back to non-secure connection in case pk mismatch”). I can’t find any actual documentation of the protocol and it’s a custom thing built on NaCl boxes, yikes. [4] generates a keypair on the fly for a new connection to sign the box containing the symmetric key generated at the same time…
> A modern consumer GPU can crack "four random English words" in a day.
This is just completely false.
Fidelix 19 hours ago [-]
Did you even check what hash they are using?
applfanboysbgon 17 hours ago [-]
SHA-256. Did you? My point isn't constrained to this exact service, though. My point is that the XKCD-style passphrase is in general not secure. If you make a habit of using it, assuming that the service in question will take care of securing it super duper safely on your behalf, you will get bitten when a service doesn't do this.
XorNot 17 hours ago [-]
If character class constraints are enforced then that password becomes incredibly strong though - e.g. require a capital, and special characters and a number.
Plenty of sentence passwords meet that but now the probability space has exploded.
LoganDark 9 hours ago [-]
> the XKCD password scheme being proposed there
I proposed replacing the ridiculous checklist with zxcvbn.
I agree xkcd's scheme is a bad example and distracts from the real point, which is that a checklist is fundamentally bad, not that some specific password format should be allowed.
throwaway27448 21 hours ago [-]
> A modern consumer GPU can crack "four random English words" in a day.
Sure, if you can rely on users using a specific format. The joy of the xkcd technique is you don't need to tell other people what yours is.
But, people just aren't going to remember strings of gibberish. Expecting users to do this is just silly.
applfanboysbgon 21 hours ago [-]
The "joy of the XKCD technique" is that it prescribes a specific format millions of people will use, and it's so trivial to break that you can throw it into your cracking algorithm at virtually no cost.
If you were willing to use a bespoke, more secure variation of it, you could include a capital letter and a number rather than filing an issue on a repo insisting that you be allowed to use exactly the insecure variation.
throwaway27448 20 hours ago [-]
I think you are both greatly overestimating the number of people who remember what xkcd is and their willingness to do what a webcomic prescribes uncritically. What you describe as "bespoke" is likely the normal way someone might use it: with their chosen format.
For instance, I have multiple separator characters, multiple beginning and end characters, and numbers places at specific places, and arbitrary casing. That may be overkill but it's still essentially the xkcd format.
applfanboysbgon 20 hours ago [-]
> I think you are both greatly overestimating the number of people who remember what xkcd is and their willingness to do what a webcomic prescribes uncritically
XKCD's youtube videos get millions of views each. It is a very popular comic, and that particular password advice has spread beyond it.
> What you describe as "bespoke" is likely the normal way someone might use it.
And yet the issue in question is about someone who refuses to adapt it.
strbean 20 hours ago [-]
The user in question might have their own personal format that doesn't meet the RustDesk constraints. E.g. they may not use capitals, but use `word1$word2&word3@word4`.
Tayloring your format to the requirements of each site is a step towards defeating the utility of the system; now the user needs to remember some arbitrary hard-to-remember details for each password.
Although having a standard format for your passphrases across sites removes some of the advantages over just using the exact xkdc format (when one of your passwords is compromised and attackers can guess your custom format elsewhere now)...
throwaway27448 20 hours ago [-]
This is a non-issue with a password manager
XorNot 16 hours ago [-]
"Look up - there's 7 polywaffles in the sky!"
That phrase will pass almost every single password requirement, and no one is going to crack it by brute force dictionary attack.
throwaway27448 20 hours ago [-]
> And yet the issue in question is about someone who refuses to adapt it.
Does this person exist?
pineapplepizza6 15 hours ago [-]
[dead]
j_aime 14 hours ago [-]
I've been using rustdesk for years. I love it.
GlacierFox 19 hours ago [-]
Does this work without a screen on the remote machine? I have a server with SSH access if like to view for bespoke tasks now and then.
3 hours ago [-]
aspbee555 20 hours ago [-]
RustDesk works so much better and easier than vnc and other vendored solutions. No need to open ports, no need for vpn, it just works, no account BS, no vendor lock in BS.
I have used RustDesk for years, also run my own lookup/relay server so I do not need to rely on the public lookup server
While everything else gets enshitified RustDesk just keeps getting better
wartywhoa23 20 hours ago [-]
Definitely a lifesaver for tech support!
And the fact it's self-hosted is priceless.
SubiculumCode 19 hours ago [-]
Without self-hosting, how can I be sure that the middleman server layer won't affect my security?
zuzululu 19 hours ago [-]
wonder if it can support gaming and sounds too ?
akaykhan 10 minutes ago [-]
[flagged]
superkuh 12 hours ago [-]
This depends on libdrmtap "A (LLM AI coded) zero-dependency C library for DRM/KMS framebuffer capture on Linux" https://github.com/fxd0h/libdrmtap which doesn't actually seem to give remote access so much as remote access to the screen visuals. The other hard parts of remote access on the various waylands are still left up to more proprietary, per wayland compositor, custom extensions not in wayland core or exposed by the kernel. That is, not universal.
maxlin 6 hours ago [-]
Rustdesk has been a nice alternative/replacement for Anydesk, I use both 10+ times a day. Was having issues getting any remote desktop working yesterday with my Orange Pi 5, so good to see this very issue being solved. It does sound like Wayland has some design issues in how difficult this has been though. If a piece of software needs to do this, it shouldn't be so obnoxiously difficult, as I've heard this is a major PITA with OBS too.
husky8 13 hours ago [-]
Kvm companies shutter
pudgywalsh 18 hours ago [-]
They've been working on Wayland for what, nearly 20 years now, and it still has not reached feature parity with 1996 Windows NT?
cryo32 17 hours ago [-]
To be completely fair it's 2007 Windows NT they haven't caught up with.
magicalhippo 15 hours ago [-]
Windows NT4 (1998) had officially support for Terminal Services, so RDP[1].
That's kinda the point though, no? Wayland creators threw out the baby with the bath water, and thus it's taken ages to get fairly basic functionality, like what RustDesk announced here, working on the Linux Desktop.
cryo32 6 hours ago [-]
Yes that. Also it's quite a mess, which makes stuff like this much harder than it should be.
I'm still using X11 on my Linux box and am quite happy with it.
pineapplepizza6 15 hours ago [-]
[dead]
7bit 21 hours ago [-]
Okay. The feature I'm waiting for is the self-hosted web client.
Kelteseth 20 hours ago [-]
Funny that the video in the blog post still needs X. I will see myself out....
Fidelix 19 hours ago [-]
No need to announce
senectus1 11 hours ago [-]
oooh any chance it can be integrated into guacamole?
zuzululu 21 hours ago [-]
so this means I can be on a vacation, turn on my ubuntu desktop remotely, login and control it ? I have a strong need for this as my desktop is also a server
amelius 21 hours ago [-]
You can already do this with VNC or Xpra (the latter is a screen/tmux for graphical applications).
zuzululu 19 hours ago [-]
so whats the edge that rustdesk that offers ? gaming ?
yjftsjthsd-h 15 hours ago [-]
Performance, at any rate. VNC is fine but can be laggy.
Fnoord 11 hours ago [-]
I ran Wayland over Waypipe over WireGuard (Tailscale, in my example), and it worked very well. That over a 1 gbit fiber... shared with whole building/office. In fact, I ran remote AOSP this way.
Now, VNC might actually also work very well over WireGuard or SDN based on WireGuard. Since VNC was designed for LAN, in a time where LAN was the common use case.
For me, low latency is from the past. I always got either wired, Wi-Fi, LTE, or Starlink.
amelius 5 hours ago [-]
VNC works fine.
VorpalWay 21 hours ago [-]
For a remote Linux system you can also do a lot over plain SSH as well.
shevy-java 18 hours ago [-]
Is Wayland finally feature complete to xorg or are they still refusing to?
ChocolateGod 4 hours ago [-]
Wayland does not implement a print server or drawing API no.
Not true at all. RustDesk connections are fully encrypted even when using a self-hosted server. What is not supported are encrypted direct endpoint-to-endpoint connections not going through a server at all, because that's not a use case RustDesk is designed for. The feature is intended for testing, which is why it's disabled altogether by default.
If your use case is focused on direct connections without a server, RustDesk is the wrong tool for the job. It's an open-source self-hostable analog to Teamviewer, the server is inherent to the design.
VNC does not, for example, and you need to set up port forwarding through ssh. X2Go does....but I don't trust it terribly. So, RustDesk..what do you need to wrap it in? ssh?
I really want a better solution that VNC/X2Go..especially since I use hardware accelerated apps that need to use VGL to operate correctly (but underwhich operation is quite buggy).
I think such setups are at least as secure as having tailscale in front of it and they are web standards conform. I dont need a client app like tailscale, I can just use my normal browser and internet conn.
I always make sure envoy/all other apps are on the latest security patcb anyways.
What all user-facing software should have is a minimal-overhead connection option to improve performance inside user's tunnel of choice.
Pure QUIC gets blocked easily, SSH requires wrapping, even Tailscale mimicry is basic and they still ignore simple protocol improvements available.
basically an embedded tailscale.
Please just make it work seamlessly with my existing SSH credentials. Like SFTP.
Nothing prevents RustDesk from implementing TLS TOFU (see e.g. Gemini Protocol), which offers the same security guarantees as SSH TOFU.
But yeah, you could also make use of your ed25519 ssh public key as client certificate and accept based on fingerprint like ssh
"We don’t plan to implement this directly, as it is designed for test initially, we do not recommend using RustDesk with a public IP alone, that's also why direct ip access is turned off by default.
For security and privacy, we strongly advise using it in combination with a VPN—the VPN tunnel already provides end-to-end encryption, so additional encryption within RustDesk is unnecessary in that setup.
That said, we welcome community contributions! Many users express interest or share feedback—but very few take the next step and submit a pull request. If you’re able to help, your PR would be greatly appreciated! "
hate this. "we don't want this on principle... but we would accept it if you did it yourself!" just say you don't feel like it! don't act like it's a bad thing except for when somebody else does it for you, that's just dishonest.
What is there to hate here, I don't understand?
They first provide a concise overview of the current situation. Then how they personally feel about it. Finally they leave it open and end it with "If you want it badly enough, we'll accept outside contributions for it".
They're not saying that they're against the feature in principle, they're saying that they themselves don't plan to spend time implementing it, but if users really want it and contribute the feature itself, they'll be happy to maintain it once merged.
This seems like the ideal solution? What would be better here?
Factually incorrect. If you self host a relay/coordination server - encryption works as documented.
[1] https://github.com/rustdesk/rustdesk/issues/3714#issuecommen...
1 - https://github.com/rustdesk/rustdesk/issues/3714#issuecommen...
>RustDesk is a single open-source application (AGPL) with its own architecture. Clients connect outward to an ID/rendezvous server, which brokers a peer-to-peer or relayed session. Per the RustDesk documentation, traffic is end-to-end encrypted (built on NaCl)
Edit: i appreciate the explanations; thank you.
It consumes, on the client, 800%-1200% CPU AND 8%-10% GPU decode on my NVIDIA card. I opened a bug and they transformed it into a discussion, without any response.
How dare they.
That's what made me uninstall RustDesk.
If you're experiencing a bug that pegs you're CPU, fair enough. But RustDesk is video-stream based (in many formats), while TightVNC is image(-region) based. This is actually similar to why you watch movies in, say, H.264 rather than Motion JPEG¹.
¹=similar, not same.
No, thanks.
As they say: "the proof of the pudding is in the eating". So I used both applications and TightVNC is just a lot better, particularly in constrained remote VMs.
Docs are cheap, performance is what matters.
https://tigervnc.org/
> TigerVNC was originally based on the (never-released) VNC 4 branch of TightVNC.
It seems to be popular, that’s why I’m asking.
Funny thing is that the document where they'd explain their reasoning to fork TightVNC returns 404.
If you already use VNC this is not something for you
https://github.com/UNITRONIX/BetterDesk#prerequisites
I just glanced over the repo, deployment and comments around it and it seems like unconstrained AI slop or at least the project seems to be led by a person which can't use AI effectively for the purpose of software engineering.
Not that RealVNC is VNC. However, enshittification has clearly taken another scalp.
How does RustDesk compare with that?
GNOME Remote Login runs an RDP server, and RDP indeed isn’t fast enough, although is TLS-encrypted.
There's VNC and VNC-likes (rdp, steam link, rustdesk, etc) which all do similar: stream a screen, allow remote keyboard/mouse, maybe redirect sound, maybe use video encoders.
There's also things like deskflow, which allow you to use one keyboard/mouse and control with multiple machines.
https://files.catbox.moe/d5ztxi.jpg
Does it framebuffer grab the current session and inject input events?
It entirely sidesteps the problem of X vs Wayland, although this does mean the unattended physical seat can be physically taken over while you are remote.
The basic idea is to enumerate the CRTCs to get the relevant framebuffer plane resources and then use dma-buf exports to pass them to an encoder. Reality can be a bit more complex due to the existence of overlay planes but that's the idea.
https://github.com/LizardByte/Sunshine/blob/master/src/platf...
So far it's only feature I found missing compared to proprietary solutions.
I will note that the XKCD password scheme being proposed there is, in fact, completely insecure. A modern consumer GPU can crack "four random English words" in a day. You can argue that it's the user's choice to be allowed to use insecure passwords, but arguing that that scheme is actually secure is just wrong.
Let's run the math:
EFF's long wordlist[0] (the one used by Bitwarden's passphrase generator) has 7776 words, which is about 13 bits of entropy per word (log2(7776) = 12.92). A 4-word passphrase then has 51.7 bits of entropy, meaning there are 2^51.7 possible options for the passphrase, which is 3.66E15 possible options.
For reference, a password with all completely random characters and symbols, for each character you have 70 possibilities, or about 6 bits of entropy. An 8-character completely random password (no words or the usual patterns) then has about 49 bits of entropy, which is less than a 4 word passphrase.
About the time it takes to crack it, assuming you can test an average of 1 password per μs (you probably can't, as passwords are usually stored using key derivation functions[1][2] with work factors tweaked for current hardware) it would take ~116 years to crack it. I usually see 5-passphrase recommended nowadays, which would multiply the required effort by 7776.
Sure, a 16-character random password has a lot more entropy, but it's also a lot harder to interact with (reading, comparing, typing) if you end up needing to, and it's still easier to crack than an 8-word passphrase.
[0] https://www.eff.org/dice [1] https://en.wikipedia.org/wiki/Key_derivation_function [2] https://cheatsheetseries.owasp.org/cheatsheets/Password_Stor...
You are not (usually) in control of what services use to encrypt your password, and most software in the world uses a basic hash like MD5 or SHA-256 rather than a key derivation function, meaning you are wildly, wildly underestimating the number of guesses per second here. To suggest that passwords are "usually" stored like that is just wildly incorrect. Only well-educated, high-quality engineers who care about security, a vanishingly tiny portion of the industry, even think about things like this. I still run into government services that e-mail me my password in plaintext.
---
Sorry, but edit attaching a reply to another comment here because HN rate-limits me:
> This is just completely false.
My mid-tier laptop GPU can crack the usual wordlist in under 12 hours against MD5 (46 billion guesses per second), or about 3 days against SHA-256 (6.5 billion guesses per second). A desktop 4090 would likely crack SHA-256 in under a day (on average).
I make claims with confidence because I am speaking about facts. Why are you making claims with confidence when you have no idea what you're talking about? The quality of discourse on this site is so, so low.
> A modern consumer GPU can crack "four random English words" in a day
What you're saying now:
> My mid-tier laptop GPU can crack the usual wordlist in under 12 hours against MD5 (46 billion guesses per second)
So your response to the plain mathematical fact that, no, your consumer GPU cannot crack "four random English words" in a day, is that your consumer GPU can crack four words chosen from a list of a few thousand, in a day. Followed by personal attacks. Okay.
Trying to pick a strong password assuming the service you’re signing up for doesn’t care about security is pointless. You’ll be pwned regardless of what you do.
For passwords? Where do you get that information?
Even if only 10% of the services you use use MD5/SHA-256 (although I certainly would expect it to be >50% if we could do a large-scale study), why accept your password being easily compromised 10% of the time?
Those services using md5 aren't doing important things, I hope, so the password for that service hardly matters.
If it does matter, then user long passwords/phrases and rotate every time (or find a better supplier).
Let me do just that!
This is a networked service. You send your password (or a hashed form) to it, and it validates it. You don't have the local hash to bruteforce it offline.
Even if we only consider the top 10k english words, it's 10000^4. It's going to take years to bruteforce this over a network because you'll go through so many rate-limits, cooldown periods, and outright bans that it's questionable whether it's even possible.
Hopefully neither. But given everybody involved in VNC seems to be the same batch of clueless morons who built all those PHP web forums twenty years ago with MD5 as the password hash, who knows what they cobbled together. Maybe an expert can chime in about what actually happens here?
Yes, for a sensible scheme this can't work.
I would recommend never using this software.
[1]: https://github.com/rustdesk/rustdesk/blob/7aa98d43cf1962a7a2... [2]: https://github.com/rustdesk/rustdesk/discussions/8392 [3]: https://github.com/rustdesk/rustdesk/blob/7aa98d43cf1962a7a2... [4]: https://github.com/rustdesk/rustdesk/blob/7aa98d43cf1962a7a2...
This is just completely false.
Plenty of sentence passwords meet that but now the probability space has exploded.
I proposed replacing the ridiculous checklist with zxcvbn.
I agree xkcd's scheme is a bad example and distracts from the real point, which is that a checklist is fundamentally bad, not that some specific password format should be allowed.
Sure, if you can rely on users using a specific format. The joy of the xkcd technique is you don't need to tell other people what yours is.
But, people just aren't going to remember strings of gibberish. Expecting users to do this is just silly.
If you were willing to use a bespoke, more secure variation of it, you could include a capital letter and a number rather than filing an issue on a repo insisting that you be allowed to use exactly the insecure variation.
For instance, I have multiple separator characters, multiple beginning and end characters, and numbers places at specific places, and arbitrary casing. That may be overkill but it's still essentially the xkcd format.
XKCD's youtube videos get millions of views each. It is a very popular comic, and that particular password advice has spread beyond it.
> What you describe as "bespoke" is likely the normal way someone might use it.
And yet the issue in question is about someone who refuses to adapt it.
Tayloring your format to the requirements of each site is a step towards defeating the utility of the system; now the user needs to remember some arbitrary hard-to-remember details for each password.
Although having a standard format for your passphrases across sites removes some of the advantages over just using the exact xkdc format (when one of your passwords is compromised and attackers can guess your custom format elsewhere now)...
That phrase will pass almost every single password requirement, and no one is going to crack it by brute force dictionary attack.
Does this person exist?
I have used RustDesk for years, also run my own lookup/relay server so I do not need to rely on the public lookup server
While everything else gets enshitified RustDesk just keeps getting better
And the fact it's self-hosted is priceless.
[1]: https://en.wikipedia.org/wiki/Remote_Desktop_Protocol
I'm still using X11 on my Linux box and am quite happy with it.
Now, VNC might actually also work very well over WireGuard or SDN based on WireGuard. Since VNC was designed for LAN, in a time where LAN was the common use case.
For me, low latency is from the past. I always got either wired, Wi-Fi, LTE, or Starlink.