The article is short; go read it then come back and delete.
$ getent services gopher
gopher 70/tcpalso `curl -v http://127.0.0.1:gopher/` gives error message
* URL rejected: Port number was not a decimal number between 0 and 65535
* Closing connection
curl: (3) URL rejected: Port number was not a decimal number between 0 and 65535
is it http or gopher? :)
i just wanted to make the point that even if you have service names in /etc/services, it is not possible to use that names easily to host/access http(s) services.
nc is for generic connections and handles it well.
if you host an http(s) service on port 11111 you can reach it with url http://127.1:11111, but url http://127.1:vce/ would not work in most software.
$ grep 11111 /etc/services
vce 11111/udp # Viral Computing Environment (VCE)
vce 11111/tcp # Viral Computing Environment (VCE)You can still publish port numbers along with addresses in DNS though (SRV records).
It's never.
After decentralisation we always see decentralisation. After a period of growth, a decline will follow. After the vibe coding hype, consolidation will follow. After rain comes sunshine.
People shit-talk container orchestration systems like Kubernetes, but if anything they greatly simplified (if not completely eliminated) the need for this sort of network bookkeeping.
Want to run another webserver instance or whatever on your computer? Get the OS to allocate a new IP for it. Ports be damned.
Could be implemented in a backwards compatible way by requiring all IPv6 TCP/UDP traffic to use a fixed port number.
Yes, that's why I said I know it was mixing of layers.
However ports are a layer violation in a strict sense, introduced as a workaround because there was no easy way to just add thousands of new IPs to a single host back in the IPv4 days. No need to continue a workaround that causes grief on a daily basis.
E.g. "telnet localhost ssh" takes you to port 22 (not the default 23 for telnet). This works because /etc/services maps "ssh" to "22".
If you're sick of remembering port numbers, create some entries in your /etc/services.
Of course, only programs which use getservbyname to resolve port numbers will accept your names.
i dont know why people keep insiting on that file while there are perfectly fine commands to pull from your boxes what is holding what port.
that is all besides the point though if you look at what you should be doing and keeping all this information in some kind of asset management system from which you can deploy things (which is kinda what k8s and docker etc. try to do (miserably)).
unless you are binding stuff to random ports on random boxes there is no need to do any of it at runtime and you can just consult your bookkeeping (for which etc services lacks a lot of details to use...)
I find out what all my local servers are by `cat /etc/hosts`, because I put them in there. They run using an entry in the nginx config.
For short-lived stuff I don't even bother with that, I just use `whatever.localhost`.
If there was no LLM, author would have put a little more thought into this, maybe did a google search, and realised that all he needed were two shell scripts.
The more you use LLMs, the less you actually think
> The real annoyance is that it wasn’t just one machine. It was layers.
> I wanted a simple launcher for all the things that aren’t traditional desktop apps. Not Finder, Alfred or Raycast.
The entire damn article is like this - why would I trust software to run on my local machine when it was written by someone who did not even take care writing a blog post? How much care would they have possibly put into reviewing their vibe coded slop if they couldn't even bother to review their blog post?
[0]: https://idiallo.com/blog/say-no-to-localhost3000-use-custom-...
I've been wanting something like this for local Dev, but I think more:
Per user DNS.
So if the process doing the lookup is my own then redirect to the named service.
That would work if your goal was to route traffic to localhost.
What if it isn't?
There are reasons why the likes of example.com exists.
> So I built local.vibe — a friendly dashboard and local .vibe hostname for every local web app on your Mac. No more localhost:3000 vs localhost:5173 roulette.
> The whole thing communicates over a Unix socket acting as a reverse proxy. No external services, no accounts, no telemetry.
We’re discussing a tool that is designed for – and is only capable of – routing traffic to localhost. It’s perfectly reasonable to point out that there’s an easier solution for this use case.
example.com, and the reserved TLD ".example", exist for technical documentation and writing. If you are writing a comment on HN, or a curriculum for a networking class, then you can discuss "foo.example.com connects to bar.example.com" or "Let's hypothesize about two offices called accounts.example and human-resources.example"
The "example" domains are never supposed to reflect anything that is actually deployed onto LANs, or test labs, or the Internet, current situation notwithstanding.
https://en.wikipedia.org/wiki/.example
There are, likewise, IPv4 and IPv6 ranges that are reserved to be used in documentation. Not the 192.168.0.0/24 or 10.0.0.0/8, but separate ranges that writers only write about, and are never deployed, not even in private.
localhost is only ever going to be the loopback interface, never across a network: https://en.wikipedia.org/wiki/.localhost#Conventional_use
See also: https://en.wikipedia.org/wiki/.test
The latter article lists foreign-language TLDs which serve the same purpose.
Some proposals are described here: https://en.wikipedia.org/wiki/.home
I've also come across projects using a public DNS record that points to 127.0.0.1 (something like localtest.me?). IMO that's way worse than using .localhost since you're trusting some rando not to change the DNS records and exfiltrate your meant-to-be-local traffic.
As for 127.0.0.0/8 in the public DNS: https://utcc.utoronto.ca/~cks/space/blog/sysadmin/HowNotToDo...
As for localnet and localhost in general:
https://utcc.utoronto.ca/~cks/space/blog/sysadmin/LocalhostI...
https://utcc.utoronto.ca/~cks/space/blog/web/LocalhostSurpri...
".vibe" is not a TLD. It is not a registered TLD; it is not a reserved name. It isn't a domain at all. Go ahead, do a WHOIS lookup. Anyone who attempts to use such gibberish, even in documentation, deserves to be rudely surprised, someday in the future.
Not too long ago I had a similar issue and solved with that.
Granted no fancy UI to start and stop things but is it really needed?
Tbh this is not a single binary you need dnsmasq go and other things
From a brief glance over the code I like the approaches I see. Using the `/etc/resolver/` mechanism is a new trick to me!
The interesting part to me isn't the port numbers, it's the automatic service start/stop, including idle route shutdown.
> The real annoyance is that it wasn’t just one machine. It was layers.
173 looks like ITE
5 in roman numerals is V.
If you type "8483" on T9, your phone may offer "THUD" or "TITE" or all three, as choices.
But with a normal telephone keypad, if you dial, e.g. "(800) 555-VITE" then you will always dial "8483".
https://en.wikipedia.org/wiki/Phoneword
Also, a service port is always qualified by its protocol. There are separate port namespaces for each IP protocol that uses ports. "8483" is not a service port, until you spell it out:
8483/tcp
8483/udp
8483/sctp
8483/dccp
A TCP stream, for example, consists of a tuple:
src:port1 dst:port2I am also sick of handling port numbers - I end up allocating them on a schema to different services, so for testing I can spool any VM/service combination and avoid crossover. But if I want the same service twice, ah...
It always fascinated me that ports don't have any kind of textual resolver, so you can bind to `:1234` and also say "please also accept `:foobar`". But that would itself require some kind of "port resolver" on a device, and that's another service to break and fix :)
getservbyname(3)
I Got Sick of Remembering Port Numbers
https://gregraiz.com/blog/local-vibe/