here I don't post like I used to, everything turns into an argument now.

Yeah, well ... up yours! <G>

... Press any key to continue or any other key to quit

--- Renegade v1.40/DOS
* Origin: The Titantic BBS Telnet - ttb.rgbbs.info (21:1/144)

On 24 Mar 2026 at 09:00a, deon pondered and said...

The rate limiting will impose the following limits:

* Maximum 2 concurrent connections,
* Maximum 5 connections in a 60 min period.

Will be interested to see how this goes.

At my end I have not seen any significant issues of concern in this regard, from memory the highest frequency of polling I've seen was around every 5 mins and to me that's fine..

I don't mind if that level of activity occurs and have enjoyed the slightly delayed experience of an almost real time chat with others around the globe via echomail which is facilitated by systems crashing traffic to each other as it lands in their tossers..

Kerr Avon [Blake's 7] 'I'm not expendable, I'm not stupid and I'm not going' avon[at]bbs.nz | bbs.nz | fsxnet.nz

--- Mystic BBS v1.12 A48 (Linux/64)
* Origin: Agency BBS | Dunedin, New Zealand | agency.bbs.nz (21:1/101)

Re: Re: Hub 3 rate limiting
By: Avon to deon on Sat Mar 28 2026 11:52 am

Howdy,

* Maximum 2 concurrent connections,
* Maximum 5 connections in a 60 min period.

Will be interested to see how this goes.

I don't mind if that level of activity occurs and have enjoyed the slightly delayed experience of an almost real time chat with others around the globe via echomail which is facilitated by systems crashing traffic to each other as it lands in their tossers..

Yeah, I'm not liking it because I do want to accept mail anytime, when somebody has mail to send - but its a few port probes and nodes that are polling <5mins and not deliverying any thing that I want to target.

I'm all in on the crash mail concept - ie: you have mail to send, send it immediately, and I dont want to send out mail immediately as well. So I may make some improvements on this if and where I can.


...ëîåï
--- SBBSecho 3.37-Linux
* Origin: I'm playing with ANSI+videotex - wanna play too? (21:2/116)

Hi Exodus,
On <Sat, 27 Mar 26>, you wrote me:

Yeah, well ... up yours! <G>

LOL!!!!!

... Failure has gone to his head.


* SeM. 2.26 * From the Dirty Shwa
--- SBBSecho 3.37-Linux
* Origin: _thePharcyde telnet://bbs.pharcyde.org (Wisconsin) (21:1/700)

On 28 Mar 2026 at 07:45p, deon pondered and said...

Yeah, I'm not liking it because I do want to accept mail anytime, when somebody has mail to send - but its a few port probes and nodes that are polling <5mins and not deliverying any thing that I want to target.

it's like you need a temp data base that tracks how many polls per IP or declared node number ... is that how it works?

I'm all in on the crash mail concept - ie: you have mail to send, send it immediately, and I dont want to send out mail immediately as well. So I may make some improvements on this if and where I can.

For me if someone sends mail in I'm more than happy to see it tossed and CRASHed to nodes that ask for that and if others on HOLD poll in then they get it when they want it..

Kerr Avon [Blake's 7] 'I'm not expendable, I'm not stupid and I'm not going' avon[at]bbs.nz | bbs.nz | fsxnet.nz

--- Mystic BBS v1.12 A48 (Linux/64)
* Origin: Agency BBS | Dunedin, New Zealand | agency.bbs.nz (21:1/101)

The rate limiting will impose the
following limits:

* Maximum 2 concurrent connections,
* Maximum 5 connections in a 60
min period.

Hey Deon,

I was just curious if this expands over to the REST API as well or if it was just for binkd sessions?

I'm not sure who else is using the REST API but I do poll that for netmail every 30 min and send out the outbound netmail to the API as it happens in realtime.

Thanks!


--- NE BBS v2.00 (linux; x64)
* Origin: NE BBS - nebbs.servehttp.com:9223 (21:3/193)

Re: Re: Hub 3 rate limiting
By: slacker to deon on Sun Mar 29 2026 04:58 am

Howdy,

I was just curious if this expands over to the REST API as well or if it was just for binkd sessions?

I'm not sure who else is using the REST API but I do poll that for netmail every 30 min and send out the outbound netmail to the API as it happens in realtime.

No, its doesnt include the REST API - not many are using it, so its not been a problem. (When it is a problem, I'll probably do something then).

And while we are on the topic, I've been doing a lot of internal optimisations, is the REST API still working OK? (Just want to make sure I didnt break it.)




...ëîåï
--- SBBSecho 3.37-Linux
* Origin: I'm playing with ANSI+videotex - wanna play too? (21:2/116)

Re: Re: Hub 3 rate limiting
By: Avon to deon on Sun Mar 29 2026 03:40 pm

Howdy,

it's like you need a temp data base that tracks how many polls per IP or declared node number ... is that how it works?

At the moment it is done via HAPROXY, so its by IP address.

I may add the feature into clrghouz, then I can take into account past sessions and transfer details (I have that data).

I think I also read that HAPROXY can also be controlled via an API, so I could even tell HAPROXY to forget what it knows about an IP address when incoming mail comes in via a poll (thus effectively reseting the counters).

For me if someone sends mail in I'm more than happy to see it tossed and CRASHed to nodes that ask for that and if others on HOLD poll in then they get it when they want it..

Yeah, me too - its the few nodes that are on autopilot that just set up their system and forget it. It looks like its down to about 4 IPs that are hitting my binkp every minute now, some multiple times each minute.

It seems the script kiddies are not working this weekend, or I just havent noticed them now :) In the 48hrs I've implemented this, its dropped about 9000 connections.

While I havent checked all the IPs that have been rate limited, the ones I have check also receive their mail successfully via CRASH :(

That said, if somebody is genuinely affected by this let me know and I'll work out something - otherwise from my view of the world, I've fixed what I set out to fix.


...ëîåï
--- SBBSecho 3.37-Linux
* Origin: I'm playing with ANSI+videotex - wanna play too? (21:2/116)

No, its doesnt include the REST
API - not many are using it, so
its not been a problem. (When it
is a problem, I'll probably do
something then).

Ok cool! Thanks!

And while we are on the topic,
I've been doing a lot of internal
optimisations, is the REST API
still working OK? (Just want to
make sure I didnt break it.)

No issues that I've seen in my logs. There was some errors about a week ago (I think) but I think all of clearing houz was down for a tiny bit.

I'll run a few netmail tests just to make sure and let you know if I have any issues.


--- NE BBS v2.00 (linux; x64)
* Origin: NE BBS - nebbs.servehttp.com:9223 (21:3/193)

And while we are on the topic,
I've been doing a lot of internal
optimisations, is the REST API
still working OK? (Just want to
make sure I didnt break it.)

Just ran some tests and it does seem like something might be amiss.

I sent two netmails to the API and both returned a successful response but I don't see them on the clearing houz web interface (I forget if they show up there or not though). Also it looks like maybe I hit the rate limit on the web or something else happened because I wasn't able to access it for a bit.

My server has since polled the API for any pending netmail and none were found. (One of the netmails I addressed to myself on my node so I think I should have received that one unless self sending like that is disabled..)


--- NE BBS v2.00 (linux; x64)
* Origin: NE BBS - nebbs.servehttp.com:9223 (21:3/193)

Re: Re: Hub 3 rate limiting
By: slacker to deon on Mon Mar 30 2026 05:23 pm

Howdy,

Just ran some tests and it does seem like something might be amiss.

Ahh, I think I found it - missed something in the optimisation work.

I'm pushing out an update, could you try again and let me know.

Thanks


...ëîåï
--- SBBSecho 3.37-Linux
* Origin: I'm playing with ANSI+videotex - wanna play too? (21:2/116)

Ahh, I think I found it - missed something in the optimisation work.

I'm pushing out an update, could you try again and let me know.

Ran a few more tests today and everything seemed to work now except for one inbound netmail I sent from another BBS seems to have never made it here. I'm not sure how their configured though so it's possible it just never sent it out.

I tested:
* Self sending to my own node from my bbs - success
* Sending to a remote BBS from my BBS - success
* Replying from a remote BBS to my BBS - success except for 1 server.

It looks like its working again (unless you see anything on your side about that missing message).

Thanks!!


--- NE BBS v2.00 (linux; x64)
* Origin: NE BBS - nebbs.servehttp.com:9223 (21:3/193)

Re: Re: Hub 3 rate limiting
By: GRiM to fusion on Thu Mar 26 2026 10:25:22

These echos do not get enough traffic to warrant polling once per minute - hard stop. Crash delivery reduces the need to poll unless you've been offline for awhile. There's no reason to poll more than once per hour.

I agree with this statement.
--- SBBSecho 3.37-Win32
* Origin: Themegabbs.ddns.net 6400 (21:4/114)