Hi Rick!

04 Oct 2021 07:02, from Rick Smith -> All:

I am curious about dupes. Ive ignored them up until now, and now that
I have time Id like to explore why they happen, is there something I
can do on my end to stop them etc? This hub gets on average 20-75 per
day where do they come from?

If you have multiple links for echos (which makes very much sense for redundancy reasons) then it is completely normal. (you WANT to get multiple copies, so that if one link fails, you will still get the message, and you simply delete any additional redundancy copies)

If not then somebody uses buggy tossers.
In my experience this is often caused by all in one solutions like a BBS that has a built in (often buggy) tosser included.

CU, Ricsi

... Never interrupt your opponent while he is making a mistake.
--- GoldED+/LNX
* Origin: A hundred thousand lemmings can't all be wrong. (2:310/31)

Hi Alan,

On 2021-10-04 07:58:53, you wrote to me:

I'm curious why you are posting from /81 which is listed with sysop
name: "Michael Pierce". While you have 3 node numbers in the nodelist
for yourself...? ;-)

My nodelist says /81 is Rick Smith. The nodelist will catch up in time I am
sure..

Your right. It was changed from 275 -> 276. I'm using the weekly in Golded. So my curiosity is satisfied. ;-)

Bye, Wilfred.
--- FMail-lnx64 2.1.0.18-B20170815
* Origin: FMail development HQ (2:280/464)

On 04 Oct 21 17:54:50, Richard Menedetter said the following to Rick Smith:

If not then somebody uses buggy tossers.
In my experience this is often caused by all in one solutions like a BBS th has a built in (often buggy) tosser included.

Which software might that be?

Nick

--- Renegade vY2Ka2
* Origin: Joey, do you like movies about gladiators? (1:229/426)

Hello Wilfred!

Monday October 04 2021 16:53, you wrote to me:

* Forwarded from area 'FIDO_SYSOP'
Hi Rick,

On 2021-10-04 07:02:02, you wrote to All:

I am curious about dupes.

* Origin: Awesome Net- Oregon FTN Hub - Portland, OR -
www.awesomen (1:105/81)

I'm curious why you are posting from /81 which is listed with sysop
name: "Michael Pierce". While you have 3 node numbers in the nodelist
for yourself...? ;-)

Im posting from the HUB that he and I ran together before he passed away earlier this month. The nodelist will update to that fact this week..Just happens to be where I am noticing the dupes.

Regards,

Rick

--- GoldED+/LNX 1.1.5-b20180707
* Origin: Awesome Net- Oregon FTN Hub - Portland, OR - www.awesomen (1:105/81)

Hi Rick,

On 2021-10-04 18:43:20, you wrote to me:

I am curious about dupes.

* Origin: Awesome Net- Oregon FTN Hub - Portland, OR -
www.awesomen (1:105/81)

I'm curious why you are posting from /81 which is listed with sysop
name: "Michael Pierce". While you have 3 node numbers in the nodelist
for yourself...? ;-)

Im posting from the HUB that he and I ran together before he passed away earlier this month.

I'm sorry to hear that.

The nodelist will update to that fact this week.

It already did in the daily when I checked. So it will be in the next weekly...

Just happens to be where I am noticing the dupes.

If the hub has multiple (up)links that is normal, and can be expected in todays fidonet (it's called redundancy! ;-)). As long as your dupe detection is working ok, there is nothing to worry about.

Bye, Wilfred.
--- FMail-lnx64 2.1.0.18-B20170815
* Origin: FMail development HQ (2:280/464)

Hi Nick!

04 Oct 2021 13:20, from Nick Andre -> Richard Menedetter:

If not then somebody uses buggy tossers.
In my experience this is often caused by all in one solutions like
a BBS th has a built in (often buggy) tosser included.

Which software might that be?

I have seen it happen with Mystic.
A node running Mystic sent me back messages with the headers of a message by me, and the body was from a completely different message.

This was the worst incident, but there were multiple smaller ones.

CU, Ricsi

... Without my ignorance, your knowledge would be meaningless.
--- GoldED+/LNX
* Origin: Mary had a little lamb... a little beef, a little ham. (2:310/31)

05 Oct 2021 09:01, you wrote to Rick Smith:

If the hub has multiple (up)links that is normal, and can be expected
in todays fidonet (it's called redundancy! ;-)). As long as your dupe detection is working ok, there is nothing to worry about.

Wouldn't it be better then, for maximum redundancy, to stop using SEENBYs, at least in nodes that have multiple fidoweb-style links *and* good dupe detection?

Carlos
--- GoldED+/W32-MSVC 1.1.5-b20180707
* Origin: Costa Blanca, Spain (2:341/234)

Hi Carlos,

On 2021-10-23 12:41:30, you wrote to me:

If the hub has multiple (up)links that is normal, and can be expected
in todays fidonet (it's called redundancy! ;-)). As long as your dupe
detection is working ok, there is nothing to worry about.

Wouldn't it be better then, for maximum redundancy, to stop using SEENBYs, at least in nodes that have multiple fidoweb-style links *and* good dupe detection?

We don't want or need "maximum" redundancy. "Good" redundancy is good enough! ;-)

And unfortunately dupe checking isn't perfect in a lot of software that is currently used. Some software doesn't even check dupes before sending mail forward to their links (only on importing to their message base). That's why I sometimes see paths like this in my dupes area:

@PATH: 387/26 396/45 280/464 203/0 230/0 150 261/38 396/45
^^^^^^ ^^^^^^^ ^^^^^^

The message this line was from already passed through 396/45 and my system. Somewhere between us seenby's were stripped. But still 396/45 forwarded it to me for the second time. That wouldn't happen if it did dupe checking before forwarding. So without seenby checking and without better dupe checking, the number of dupes would explode, and some messages would probably go arround in circles forever... So it's an interesting idea to stop seenby checking, but in the current fidonet it would cause problems...


Bye, Wilfred.
--- FMail-lnx64 2.1.0.18-B20170815
* Origin: FMail development HQ (2:280/464)

23 Oct 2021 12:51, you wrote to me:

We don't want or need "maximum" redundancy. "Good" redundancy is good enough! ;-)

[snip]

So it's an interesting idea to stop seenby checking, but in the
current fidonet it would cause problems...

Thanks for your explanation, Wilfred.

Carlos

--- GoldED+/W32-MSVC 1.1.5-b20180707
* Origin: Costa Blanca, Spain (2:341/234)