[Dxspider-support] Bad IP

[Dirk Koopman]

Gregor

Is there any reason why you can't "simply^* " use an instance of 
DXSpider as your back end? I would be happy to help if that would suit 
you. It is pretty secure (no known compromises of the code itself, nor 
by some of the questionable data it is fed). It was designed, from the 
ground up, to be a self standing internet facing daemon. It 
(deliberately) does not run as root and users are actively discouraged 
from trying to do so. Having said that, it's written in pure perl, there 
are no hidden backdoors that allows the daemon to spawn arbitrary 
processes (although potentially "long running" DXSpider commands do 
spawn processes that offload these time consuming things to another 
process(or)).

As I understood it, DB0SUE-7 used a heavily modified version of the 
master branch as a back end, which is why I could never get him to upgrade.

Dirk G1TLH

* for certain values of "simple" :-)

On 05/02/2025 19:39, Gregor Surmann via Dxspider-support wrote:
> Hi Kin,
>
> no, it's not a "caprice" to insert the IP in the PC61 message, but 
> that will requiere me to write a PC61 implementation here, instead of 
> simply using telnet. I am sure, that I've read the documentation 
> carefully, but have not seen any note about the IP beeing so important.
>
> The previous solution worked for over 10 years (probably somehow 13 
> years) as I used DB0SUE-7, no one said that there was multiple 
> callsigns with the same IP, and there were maybe 3 cases of abuse in 
> those years. Now all of a sudden it's a big problem, so I'm a little 
> bit surprised ;)
>
> Regardless of that I will have to do it probably, so define the 
> webserver as node and insert PC61 from there, with valid user IP (from 
> the http request), but that may need a few days/weeks.
>
> Btw., I understand what you mean, even with my (also) "broken" English ;)
>
> 73 de Gregor, DO5SSB
>
>
> On 05.02.25 20:23, Kin wrote:
>> Hi Gregor,
>>
>> I read you and I have the impression that you think that putting the 
>> real IP
>> is a caprice. I guess it's because of my bad English 🙁
>> Dirk has tried to endow the archaic AK1A protocol with some 
>> functionalities
>> that help sysops, and therefore everyone, to detect problems, to 
>> maintain
>> the network. This has allowed us to know that a certain node/user is 
>> or is
>> not the cause of a problem, and therefore allows us to act to solve 
>> it. When
>> data is missing, problems sometimes arise and hasty decisions are made.
>>
>> When one reads the discussions here over time (I recommend it), one 
>> can be
>> amazed at what some humans are capable of doing out of boredom.
>>
>> There are web frontends likehttps://webcluster.ure.es/  ,
>> https://www.dxheat.com/dxc/ orhttps://www.dxfuncluster.com/ who are 
>> using
>> dxspider and had no problems attaching the user IP to the spots. I don't
>> remember any of them objecting to problems when I explained the 
>> reason for
>> adding the real address.
>> Even one of them I know works with HA and it works great.
>>
>> Regards,
>>
>> Kin EA3CV
>>
>
>
> _______________________________________________
> Dxspider-support mailing list
> Dxspider-support at tobit.co.uk
> https://mailman.tobit.co.uk/mailman/listinfo/dxspider-support
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.tobit.co.uk/pipermail/dxspider-support/attachments/20250206/d34abb0e/attachment.htm>