Suddenly (without any known changes having been made on my system) I get a "bad
password" error to two of the circa 20 systems I connect to.

They both have one thing in common -- "VER Argus/3.208/ binkp/1.0"

I cannot for the life of me figure out what's wrong. When I let my Argus crash the mail (the same password since decades ago) it works perfectly well, but not from binkd.

But I also encountered the old 'expected password "-"' error when no password was entered recently. Did anyone discover what this was caused by?

Is there some setting in binkd that can cause/fix this?

Any help or suggestions would be muchly appreciated!




..

--- Paul's Win98SE VirtualBox
* Origin: Quinn's Post - Maryborough, Queensland, OZ (3:640/384)

Hi Björn.

29 Apr 16 03:15, you wrote to All:

Suddenly (without any known changes having been made on my system) I get a "bad password" error to two of the circa 20 systems I connect to.

They both have one thing in common -- "VER Argus/3.208/ binkp/1.0"

I cannot for the life of me figure out what's wrong. When I let my

Argus

crash the mail (the same password since decades ago) it works perfectly well, but not from binkd.

But I also encountered the old 'expected password "-"' error when no password was entered recently. Did anyone discover what this was caused

by?

Is there some setting in binkd that can cause/fix this?

Any help or suggestions would be muchly appreciated!

I remember seeing something like this a long time ago and it was related to the
aka order.

Your binkd has akas:

VER binkd/1.1a-65/Win32 binkp/1.1
addr: 2:203/2@fidonet
addr: 2:203/0@fidonet
addr: 2:20/0@fidonet
addr: 2:2/2@fidonet

And your 'Argus':

VER Tedious/3.211/ binkp/1.0
addr: 2:203/0@fidonet
addr: 2:203/2@fidonet
addr: 2:20/0@fidonet
addr: 2:2/2@fidonet
addr: 2:203/1@fidonet

So, I'd try to adjust the order in binkd...

'Tommi

---
* Origin: ====================================== (2:221/6)

So, I'd try to adjust the order in binkd...

Ahhhh yes of course, that may be it. I DID make that change after you pointed out that I didn't have my /2 AKA in binkp.

So basically it's your fault? 8-)

--- Paul's Win98SE VirtualBox
* Origin: Quinn's Post - Maryborough, Queensland, OZ (3:640/384)

On 28.4.2016 20:49, Bj÷rn Felten - Tommi Koivula wrote:

So, I'd try to adjust the order in binkd...

Ahhhh yes of course, that may be it. I DID make that change after
you pointed out that I didn't have my /2 AKA in binkp.

So basically it's your fault? 8-)

I didn't tell you to put it on top. :)

---
* Origin: *** nntp://fidonews.mine.nu *** Finland *** (2:221/6.0)

So, I'd try to adjust the order in binkd...

That fixed it! Thanks a million, Tommi!

But isn't that strange, really? Alas, never mind, now I'm happy again. <3




..

--- Paul's Win98SE VirtualBox
* Origin: Quinn's Post - Maryborough, Queensland, OZ (3:640/384)

So basically it's your fault? 8-)

I didn't tell you to put it on top. :)

LOL! True. I don't know what I was thinking about there. 8-)

--- Paul's Win98SE VirtualBox
* Origin: Quinn's Post - Maryborough, Queensland, OZ (3:640/384)

On 28.4.2016 20:55, Bj÷rn Felten - Tommi Koivula wrote:

So, I'd try to adjust the order in binkd...

That fixed it! Thanks a million, Tommi!

:)

But isn't that strange, really?

I think it is a bug or feature of Argus...

---
* Origin: *** nntp://fidonews.mine.nu *** Finland *** (2:221/6.0)

On 28.4.2016 21:00, Bj÷rn Felten - Tommi Koivula wrote:

So basically it's your fault? 8-)

I didn't tell you to put it on top. :)

LOL! True. I don't know what I was thinking about there. 8-)

Well, the order of akas shouldn't matter... But in this case it does.

---
* Origin: *** nntp://fidonews.mine.nu *** Finland *** (2:221/6.0)

I think it is a bug or feature of Argus...

I tend to agree. But I think it would also be good if it was mentioned somewhere in the binkd documentation that the order of the AKAs matters big time.

Alas, now I only have 1:320/119 on my "hold until..." waiting list here. I'm
sure that Andrew will eventually sort his system out once again, so I'll not even bother to ask about it in the other echo...

--- Paul's Win98SE VirtualBox
* Origin: Quinn's Post - Maryborough, Queensland, OZ (3:640/384)

Hi Bj÷rn!

29 Apr 2016 05:20, from Bj÷rn Felten -> Tommi Koivula:

I tend to agree. But I think it would also be good if it was
mentioned somewhere in the binkd documentation that the order of the
AKAs matters big time.

It does not matter, when you are using a not completely bug ridden mailer!

CU, Ricsi

--- GoldED+/LNX
* Origin: Don't confuse me with facts, my mind is made up! (2:310/31)

It does not matter, when you are using a not completely bug ridden mailer!

Obviously, since only two old Argus mailers seem to be affected.

But fidonet has always been very careful always to be backwards compatible, so such a seriously bad behaviour should at least be documented and the binkd users should be warned about it IMHO.

Well, lets hope that they now have...?



..

--- Paul's Win98SE VirtualBox
* Origin: Quinn's Post - Maryborough, Queensland, OZ (3:640/384)

29 Apr 16 03:15, you wrote to All:

Suddenly (without any known changes having been made on my system) I
get a "bad password" error to two of the circa 20 systems I connect
to.

They both have one thing in common -- "VER Argus/3.208/ binkp/1.0"

I cannot for the life of me figure out what's wrong. When I let my
Argus crash the mail (the same password since decades ago) it works perfectly well, but not from binkd.

i don't think i've seen this with my taurus system...

But I also encountered the old 'expected password "-"' error when no password was entered recently. Did anyone discover what this was
caused by?

this one i've seen and it was due to a misunderstanding of the spec or what was
being seen in RL testing... some actually place a '-' in the field and transmit
that encoded as needed... what should actually be being done is not sending a password at all or sending an encoded blank/nul message...

Is there some setting in binkd that can cause/fix this?

the only thing i know of is for the developer of the software to get their binkp implementation fixed or relaxed or such... one or two of the RAT (Radius/Argus/Taurus) mailers has a problem like this and i remember reading something about bugs in similar areas...

Any help or suggestions would be muchly appreciated!

you got the code? ;)

)\/(ark

Always Mount a Scratch Monkey

... It must be Thursday. I never could get the hang of Thursdays.
---
* Origin: (1:3634/12.73)

29 Apr 16 04:55, you wrote to Tommi Koivula:

So, I'd try to adjust the order in binkd...

That fixed it! Thanks a million, Tommi!

But isn't that strange, really? Alas, never mind, now I'm happy again.
<3

that /IS/ really weird... the presentation order of addresses should not matter
one iota... what mailer expects addresses to be presented in some specific sort
order?? what order is that?

)\/(ark

Always Mount a Scratch Monkey

... Originality is the art of concealing your sources
---
* Origin: (1:3634/12.73)

29 Apr 16 05:20, you wrote to Tommi Koivula:

I think it is a bug or feature of Argus...

I tend to agree. But I think it would also be good if it was mentioned somewhere in the binkd documentation that the order of the AKAs
matters big time.

no, they don't AFAIK... it seems to matter only to one particular version of one particular mailer...

)\/(ark

Always Mount a Scratch Monkey

... Dip yer bread, chaps.
---
* Origin: (1:3634/12.73)

28 Apr 16 23:30, you wrote to Björn Felten:

They both have one thing in common -- "VER Argus/3.208/ binkp/1.0"

I cannot for the life of me figure out what's wrong. When I let my
Argus crash the mail (the same password since decades ago) it works
perfectly well, but not from binkd.

i don't think i've seen this with my taurus system...

Apparently fixed in Radius/Taurus...

'Tommi

---
* Origin: 2001:470:1f15:cb0:f1d0:2:221:6 (2:221/6)