Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!


Would anyone be interested in this?
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.

Would anyone be interested in this?

edited January 2016 in General

I'm thinking of starting a service to help peeps with NAT IP servers, so if they want HTTPS but do not want to use CloudFlare for his nameserver, I could set up a reverse proxy to help them attain that. Let's Encrypt would come in well for the situation, but provided they trust me enough haha, which I hope peeps would, since I'm not going anywhere soon anyway.

WHY THIS???

I suggested this as an alternative to CF, for the following reason:

CloudFlare is only a web reverse proxy and currently doesn't support websockets (on free plans) too, which is why I proposed this, too, as I had to disable CF when I wanted to run my own Wekan instance.

another example

For example, if I host a Git server, and my port ranges are 12300-12321, where 12321 is the SSH port for Git, I could use this server to act as a transparent proxy (maybe with this?).

This way, I can have HTTPS reachable on 443, and SSH clone/pull/push option available on other ports, since CloudFlare does not have support for passing traffic on custom ports.

I'll be happy to have such a service going on, take it as doing some good to those who want HTTPS.

Of course, manual approval but automated setup, since some peeps might abuse it.

How does this sound?

If you have the time - do leave some reasons on why (not) - I would like to better understand what y'all think!

I see, sure. Thanks alot for the honest opinion.

Thanked by 1rokok
So...
  1. Would you be interested in such a service?51 votes
    1. Heck yes I would!
        9.80%
    2. Nah I'm fine, I rather get VPSes with dedicated IPv4.
      90.20%

Comments

  • @theroyalstudent said:
    but provided they trust me enough haha, which I hope peeps would, since I'm not going anywhere soon anyway.

    Not to be offensive but I personally would not trust a 14 year old who will be starting school tomorrow and is probably depending on pocket-money earned during the vacation to fund the server.

    Shouldn't you be focussing on your national exams coming up next year?

    Otherwise, I would say good initiative but trust is totally another issue.

    Thanked by 2angrysnarl Peroni
  • Perhaps a longer lasting contribution to the community would be a well written tutorial showing people how to set this up for themselves?

  • edited January 2016

    @Server said:
    Not to be offensive but I personally would not trust a 14 year old who will be starting school tomorrow and is probably depending on pocket-money earned during the vacation to fund the server.

    Regarding funds, I can set aside 100-125 on servers per year, so that's not an issue, but as you can see on IRC, I try my best to cut down on costs of servers for myself so I can use them for other purposes and initiatives like this.

    Shouldn't you be focussing on your national exams coming up next year?

    Not yet, but I know when to start preparation. The first half or quarter of this year I can still be pretty active in code, since its nothing that serious yet.

    I still have a Co-Curricular Activity where I have to teach juniors code, so yeah.

    Otherwise, I would say good initiative but trust is totally another issue.

    Ah, I get the issue on trust, that was a thing that I thought would come up when I was thinking of starting this initiative.

    edit: this is for partially learning and partially something I want to do to give back to the society, so I don't expect people to put production stuff on my service, too.

    Thanks for the honest opinion! I really appreciate it.

    EDIT: For now, I'll take it as I shouldn't start and operate a service like that on my own - maybe I could get sponsors, or even open source the code so different people can operate this.

    also, if this gets < 50% of a "yes", I obviously wouldn't operate this until I have more funds or get older (a year or two maybe?), if that's an issue.

    @tehdan said:
    Perhaps a longer lasting contribution to the community would be a well written tutorial showing people how to set this up for themselves?

    That could work too.

  • theroyalstudent said: EDIT: For now, I'll take it as I shouldn't start and operate a service like that on my own - maybe I could get sponsors, or even open source the code so different people can operate this.

    I say go for it. You can pretty much ignore the whining about the trust aspect, I don't think it's a big problem. When I tunnel data clear in clear through a service, I just assume it is going to be compromised anyways; this is not different for CF.

  • @singsing said:
    I say go for it. You can pretty much ignore the whining about the trust aspect, I don't think it's a big problem. When I tunnel data clear in clear through a service, I just assume it is going to be compromised anyways; this is not different for CF.

    I find this true - so yeah. I might do this, but I have much more commitments in real life to take of, so I'm probably gonna write the code by June 2016.

  • The service as it is just not appealing enough as it isn't really that hard to do something like this maybe a guide could be good for non technical incline person to do it for himself.

  • @masterqqq said:
    The service as it is just not appealing enough as it isn't really that hard to do something like this maybe a guide could be good for non technical incline person to do it for himself.

    Look at LET. Youve got people asking debian vs CentOS. It is difficult for some anyway.

    I'd be happy to donate some resources and some code to the project

  • @masterqqq said:
    The service as it is just not appealing enough as it isn't really that hard to do something like this maybe a guide could be good for non technical incline person to do it for himself.

    Yea, it's easy to do, but this isn't for just one person, but the community in general.

    I plan to be making a web interface for this, which is why I decided to ask here first.

  • Nothing wrong with the initiative per se, but it's one more thing that can go down.

    The downtime is multiplied (if the proxy or the webserver is down, the website is down). Wherest paying for a dedicated IPv4 means having to worry about one only.

    Only LES like VPS would benefit from this, but those already come with a local on-site reverse proxy that adds no risk of downtime, latency or trust issues.

    To be brutally honest: this is a bad idea with no respectable future. BUT don't stop there! Keep thinking of new ideas and who knows one day you come up with a stellar idea :D

  • @theroyalstudent friend save your funds and time for something else :)

    Thanked by 1doghouch
  • @Hidden_Refuge said:
    theroyalstudent friend save your funds and time for something else :)

    Funds, nah I'm fine, I plan to be saving up some money this year.

    Time? True, I still have alot of other things to do!

  • Good idea asking before diving in and including a poll, the results of which speak for themselves, I think…

  • I recon this would be good with DDoS protection actually might be a good idea

  • IPv6 support is a must. End to end SSL would be cool if it could be done, then only the client is in control of the private key and not you.

  • Could use something like SNIproxy to proxy the request along, maintaining the initial SSL connection. That way like @linuxthefish said the customers would be in control of the SSL private key.

    Though because it's not terminating the connection it has no way of inserting the foreworded for header.

Sign In or Register to comment.