New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
Would you be able to PM it to me? I wish to deploy this for LET/LES users only for free, and wont compete with you. Ideally i'd like to deploy the most compatible version with future updates as well from your git.
I just edited agent.js for now and changed https to http.
Seems to be used for internal calls only anyway, so looks fine too me.
At some point I will switch it to https.
https works https://1.neoon.net/
However the bot page is kinda broken.
DNS Record management supported also.
![](https://i.imgur.com/gP44rGa.png)
Certs still stored locally in the DB, so you can just reuse them again, neat.
![](https://i.imgur.com/SIVzQDw.png)
What is broken about it?
Great work setting it up so fast. told you the DNS panel is amazing![:) :)](https://lowendtalk.com/resources/emoji/smile.png)
It works now, I switched it back to argon2.
Might be that I modified agent.js before, I should switch the rest to https internally.
Hmm, I use sha256 at it works perfectly fine, argon2 also worked. So would be weird if that would be the issue. Hope you get it working![:heart: :heart:](https://lowendtalk.com/resources/emoji/heart.png)
It was more like a stuck, stale haproxy config.
It should not give you "Client sent an HTTP request to an HTTPS server."
Rather it should give you something like what it is currently.
"Error 1001" since I don't have that domain on CF but just point it to 1.1.1.1:80
https://1.neoon.net/
Like what it does right now.
I restarted haproxy but that won't fix it.
So I ended up rebooting the entire machine.
Worked.
I had 2 different backends added to a domain.
I clicked delete on one, booth vanished.
This results in a stuck haproxy config.
I checked /maps to see what haproxy has been configured with, could not find the stuck config there.
If you restarted haproxy twice it would have worked. Its a bug related to backends.map. Dataplaneapi is automatically syncing it to disk and this breaks the logic in one of the lua init functions. I forked dataplaneapi to allow excluding some maps (which is also important to prevent syncing of huuuge maps like geoip). See here: https://github.com/fatchan/dataplaneapi/blob/master/configuration/map_sync.go#L48
(Looks like i havent pushed my change that adds backends.map to that list, lol. Will update my fork tonight.)
You forgot to mention that the cluster config is inside the mongodb.
The variable in the .env file never gets read.
Explains why my second server in the cluster never updated.
There is a cluster management for the db inside the Panel.
However you can add new clusters but you can't edit existing ones, which you need since your script reads only from the first element of the list.
Easily fixable with a smol python script though, works for now.
Is this still being worked on?
Yes it is. Im finalizing stuff, so this weekend or by monday hopefully.
Are you forking and making changes?
Gang up with @FlorinMarian
I still wait on some fixes, that can brick haproxy.
I would not use that in "production" yet.
I got it working (https://1.neoon.net), posted the guides and scripts.
But yea, still have to make a working guide for the Panel install.
@FatGrizzly It looks good but I can say that the ETA is useless because both on Desktop and on mobile devices the time increases and decreases until the puzzle is solved, I think it would be better to replace the message with something like "We will immediately connect you to the desired page.. ."
Any updates? I would deploy it on my server to test it how it works
https://gitgud.io/fatchan/haproxy-protection/
Also the control panel: https://gitgud.io/fatchan/haproxy-panel-next/
You can dm me for any help or questions :^)
Oh, it doesn't seem to welcome me, shouldn't there be a selection box here to verify that I am human? It directly refused my access
![](https://i.postimg.cc/pVZpR1R6/Screenshot-20231211-184944-Firefox-Nightly.png)
Have you checked that you used a different browser like FF or a different chrome browser? Incognito mode?
Been lurking on this thread for a while and still configuring mine on a LXC container on my homelab anyways. I could have misconfigured something in the config for it to loop fail or something. Eh anyways I'll try to fix that later
Sorry if this has been answered already, but does this service aim to hide web host IP addresses in the manner of Cloudflare's DNS proxies? (I know this might be a dumb question but I'm asking anyway.)
It would be helpful to know whether private browsing modes are allowed. Is incognito browsing possibly forbidden here?
My workflow depends on accessing web-based services within the guest mode of a Chromebook.
Its essentially a reverse proxy, so yes.
I still got my demo running https://1.neoon.net/ it points to 1.1.1.1 but you can't tell if it wasn't a CF error site.
Is this link supposed to demonstrate a failure of access? That's what it looks like every time I follow it. Based on your IP address provided, I'm guessing it's a little like pointing to 192.0.2.1.
I fucked around with gdnsd, it does point to 1.1.1.1 if it can't resolve the origin IP.
Especially if you request via IPv6, the .mmdb does not have any IPv6 entries.
The fallback is to 1.1.1.1
edit: Should be disabled, however probably takes time to clear
Some countries are blocked on my instance
Yes
They are allowed