Resolvers temporary rate limiting

[Saleh-Mumtaz]

Hi. Thank you for your great efforts! This app makes using dnstt far more convenient.
There are a few features I wanted to suggest, although I believe debugging may be the most important priority right now.
I see a lot of “ipv6 … rejected locally” messages in the logs, even when the proxy is working just fine. A toggle switch to completely disable IPv6 across the app would be great, as IPv6 is often the first thing restricted in Iran. It also tends to create resolving problems.
Additionally, noizdns profiles lack an MTU (Maximum Transmission Unit) value field, even though we set the MTU on the server side, in dnstt profiles, and within the app’s settings itself.

Sometimes, I connect to a noizdns server, and it works well initially, but after some time, it disconnects. This is likely because the resolver has limited the domain for a cooldown period.

If I immediately switch to another profile with the exact same resolver, it works perfectly again. In line with Morteza Bashsiz’s video, it would be appreciated to have a random subdomain toggle in noizdns or dnstt configuration profiles. Similar to the usual DNS records setup where an A record (e.g., a.example.com) points to the server IP and an NS record (e.g., ns.example.com) points to the A record, in this case, we would also need a wildcard NS record (*.ns.example.com) pointing to the A record. Cloudflare allows this, but I couldn’t create such a record in Arvan Cloud.

If some resolvers start limiting the root domain, I hope you can find a way to support multiple active domains on a single server. Currently, only one domain can remain active in noizdns. I’m unsure if this is technically feasible.

About the scanner working in the background, it seems in new androids, They have to occupy a sticky notification, by sticky i mean it can not be cleared from the bar. I suggest to show e2e progress bar with showing current scanned resolvers/total e2e resolvers number and passed number of resolvers.

Lastly, here are some logs from when the proxy was working okay, but disconnections occurred, disrupting web surfing.

Again, Thank you for creating such a great app.

slipnetv2.3.2logs

[anonvector]

Hi,

You can ignore "ipv6 … rejected locally" messages. mostly they are coming from telegram. Ipv6 is already disabled.

You can use DNS Query Size which works like MTU for DNS tunneling, it's available in the last update (see SlipNet telegram channel if you can't find it)

Rotating domains isn’t enough to bypass rate limiting. you usually need to change the server IP too.

[Saleh-Mumtaz]

Rotating domains isn’t enough to bypass rate limiting. you usually need to change the server IP too.

I have two servers, I wanted to know if is it possible for you to implement a feature in noizdns script to use one of them as a relay?
A relay with it's own domain, but all the traffic ends up in one main server. I have two DE servers, it would be nice if one of them work as a relay, two ips, two domains, less rate limiting. of course if someone is richer, then this could be implemented with more vpses and domains, up to the technical limits.

Also, those unexpected end stream happens when i am surfing web like github then everythings stops working and those are the logs i see. Is it problem to the nature of dns tunnels or it can be fixed in the future?

[Saleh-Mumtaz]

You can use DNS Query Size which works like MTU for DNS tunneling, it's available in the last update (see SlipNet telegram channel if you can't find it)

Although we have MTU value on seever side in noizdns script, yet in dnstt profiles there is query size field. Also, a lot of the times I have tried different combinations of MTU and QS values, 576 mtu with 50 QS, up to 100 QS and 1432 MTU. Sometimes mtu value also plays a roll.