HN Explorer

30 comments

[ 3.2 ms ] story [ 81.1 ms ] thread
randomerr
How do you get a query to through?
kandarpck
Tweet to @1111Resolver followed by the DNS Record type and the Domain Name. Example: @1111Resolver A http://google.com
pbalau
Not really familiar with twitter, but isn't tweeting directly to someone public? As in, everybody sees what you are trying to resolve?
drdeca
I assume this is as a joke?
randomerr
Nope, its real. Slow but real. It makes a good backup DNS so you could use it to manually update your 'hosts' files.
drdeca
I knew that it worked, I just assumed that it was made as a joke

Usually if I don't have dns access I don't have an easy way to access twitter I think? I could be wrong about this. Is this meant to be used by text messages?

ehsankia
And if you do have access to a website, there are many other sites you can use for resolving addresses.
JeremyBanks
You can use Twitter via SMS, I suppose. =P
goldfishparty
Write a short script to resolve Twitter's address. Run it on a cron to pop it into your hosts file.

Now you can get to Twitter in the case of a resolver being down. Woo!

cm2187
What I really need is DNS over Fax.
apearson
I'm sure there's a fax to twitter gateway somewhere.
xellisx
If not, give it by Monday and I'm sure some one will create an asterisk box with hylafax and have it mail it over to something to do ocr.....
petercooper
"I may have hit Twitter’s API limit." .. Yeah, Twitter's a bit like that. It might work well on Telegram though :-)
neilalexander
Given the number of people that still can't route to 1.1.1.1, this is an interesting backup!
microcolonel
Interested to see who that might be, it seems a bit nuts that that wouldn't work.
neilalexander
People who are sat behind devices configured by lazy sysadmins who copied configuration examples out of a textbook. Also people on some public Wi-Fi networks - I have noticed 1.1.1.1 being used for HTTP landing pages more than once.
arghwhat
Many. There's a lot of network configurations that accidentally use 1.1.1.1, for example for VPN endpoints.

Some AT&T users are having issues because of such fuckery.

jedberg
I mentioned this in the other thread yesterday, but my AT&T gigapower router is hard coded with 1.1.1.1 on an internal interface, so I can't use it. I can still use 1.0.0.1 though.
microcolonel
> I mentioned this in the other thread yesterday, but my AT&T gigapower router is hard coded with 1.1.1.1 on an internal interface

Wow, AT&T of all people should know better. I mean, I understand the attraction to that address for customer setup reasons, and bypassing different configured NAT address ranges (i.e. 10/8, 172.16/12, and 192.168/16 networks), but surely a customer who configures a custom address range would be capable of finding the new address.

Here's hoping that as IPv6 becomes ubiquitous, this is replaced broadly with something standard, possibly based on ND.

dingaling
And to help march towards ubiquity shouldn't we be encouraging IPv6-connected users to point to 2606:4700:4700::1111

Less junk traffic, less latency for folk behind CGNAT on IPv6-only mobile networks, no chance of collision with poorly a configured network equipment.

jedberg
I switched to IPv6, and then back to v4. IPv6 just isn't ready for prime time yet. The random timeouts waiting for it to fall back to v4 when the v6 endpoint was down was annoying enough to get me to switch back.

I'll try it again in another year.

xellisx
The last place I worked at, we 7.x.x.x network for MPLS stuff.
pbreit
Is there any difference between which address you list first in your DNS list (1.0.0.1 and 1.1.1.1)? I saw a suggestion to list 1.0.0.1 first since it likely gets much less traffic.
arghwhat
I believe they point to the same network of servers, making it exactly the same. Having two IP addresses is just custom, and makes users resilient to address issues affecting one of the addresses.

It also might help APNIC establish how much of the traffic to 1.1.1.1 is due to Cloudflare running a popular public DNS service, vs. how much is due to the address by subtracting the 1.0.0.1 traffic from the 1.1.1.1 data.

kichuku
Great! Now the whole world can see which domains you are interested in. This is supposed to be some last resort backup when all other DNS servers (8.8.8.8 and others) are down or blocked? In that extreme case of unavailability, what is the guarantee that local Govt is not involved and Twitter and/or sms to Twitter is not blocked.

Am I missing any use case here?