Hi Paul,

I've now completed 4 tests with Mystic utilizing MultiMail to send multiple messages in the same .REP packet. Each of them was successful, as each of the messages contained a different MSGID. Here are the IDs from the most recent test in fsxNet:

"@MSGID: 21:1/186 00cb2413"
"@MSGID: 21:1/186 00cb2411"

Quotes were added to hopefully not confuse any mail tossing software.


---

Dan Richter
aka Black Panther
Sysop - Castle Rock BBS
telnet://castlerockbbs.com
The sparrows are flying again...

--- Mystic BBS v1.12 A32 (Windows)
* Origin: Castle Rock BBS - castlerockbbs.com (21:1/186)

On 04/26/17, Black Panther said the following...

I've now completed 4 tests with Mystic utilizing MultiMail to send multiple messages in the same .REP packet. Each of them was successful,
as each of the messages contained a different MSGID. Here are the IDs
from the most recent test in fsxNet:

"@MSGID: 21:1/186 00cb2413"
"@MSGID: 21:1/186 00cb2411"

So on Windows with a fast processor A32 works. The problem is to get it to work on a Pi as well (since it's much slower). It doesn't work on Agency on the last test I tried. I'll have to try Castle Rock and see that it work (likely does since you've tested it) and we have a starting point.

--- Mystic BBS v1.12 A31 (Raspberry Pi)
* Origin: Mystic Pi BBS bcw142.zapto.org (21:1/145)

On 04/26/17, bcw142 said the following...

So on Windows with a fast processor A32 works. The problem is to get it
to work on a Pi as well (since it's much slower). It doesn't work on Agency on the last test I tried. I'll have to try Castle Rock and see
that it work (likely does since you've tested it) and we have a starting point.

I noticed you had stopped by to test it out. Sorry, I had the access set
higher to avoid any problems. I have lowered the access to visiting sysops,
so if you'd like to stop back and try it out. The option is "O" from the message menu. You should not be able to use it... :)

I would think that on a slower system, it would work better. The processing time would be greater to import the messages, so there would be more time between messages being processed... Just my thought process...


---

Dan Richter
aka Black Panther
Sysop - Castle Rock BBS
telnet://castlerockbbs.com
The sparrows are flying again...

--- Mystic BBS v1.12 A32 (Windows)
* Origin: Castle Rock BBS - castlerockbbs.com (21:1/186)

On 04/26/17, bcw142 pondered and said...

On 04/26/17, Black Panther said the following...

I've now completed 4 tests with Mystic utilizing MultiMail to send multiple messages in the same .REP packet. Each of them was successfu as each of the messages contained a different MSGID. Here are the IDs from the most recent test in fsxNet:

"@MSGID: 21:1/186 00cb2413"
"@MSGID: 21:1/186 00cb2411"

So on Windows with a fast processor A32 works. The problem is to get it
to work on a Pi as well (since it's much slower). It doesn't work on Agency on the last test I tried. I'll have to try Castle Rock and see
that it work (likely does since you've tested it) and we have a starting point.

Thanks Dan and Brian... well Agency is running on a Win 7 box that has a core
2 duo in it and 4 gig RAM. I don't think that's overly slow. I am not sure
the speed of the system is the contributing factor. What would be good to try and figure out are the steps / software etc. taken by each of you to get the results you got.

Brian testing on Dans system is a good idea, Dan perhaps you may want to try
a few uploaded packets on Agency - also option O on the message menu (hidden option) .. I wonder if the issue is related to multiple posts in the same
.REP packet that are replies to existing messages. I note that Dan created
two 'fresh'/non reply posts and got them to generate with different IDs.

Just some thoughts from this end :)

Best, Paul

--- Mystic BBS v1.12 A32 (Windows)
* Origin: Agency BBS | telnet://agency.bbs.geek.nz (21:1/101)

On 04/27/17, Avon said the following...

Thanks Dan and Brian... well Agency is running on a Win 7 box that has a core 2 duo in it and 4 gig RAM. I don't think that's overly slow. I am

I've got a Windows 10 64 bit system, running on a dual core with 8 gig RAM,
so not too much different.

Brian testing on Dans system is a good idea, Dan perhaps you may want to try a few uploaded packets on Agency - also option O on the message menu

I might have to try that and see what happens.

(hidden option) .. I wonder if the issue is related to multiple posts in the same .REP packet that are replies to existing messages. I note that Dan created two 'fresh'/non reply posts and got them to generate with different IDs.

That's a good point. I hadn't thought of that. I'll have to try it while replying to messages and see if I get the same results.

As far as readers, I think we're both using MultiMail 0.49.

One thing that I just thought of. I did change it to use the QWKE format as opposed to QWK. I wonder if that might make a difference...


---

Dan Richter
aka Black Panther
Sysop - Castle Rock BBS
telnet://castlerockbbs.com
The sparrows are flying again...

--- Mystic BBS v1.12 A32 (Windows)
* Origin: Castle Rock BBS - castlerockbbs.com (21:1/186)

bcw142 said....

So on Windows with a fast processor A32 works. The problem is to get it to

wor

k
on a Pi as well (since it's much slower). It doesn't work on Agency on the

las

t
test I tried. I'll have to try Castle Rock and see that it work (likely does since you've tested it) and we have a starting point.

I'm thinking as sometimes the msgids are close together, and sometimes they're far apart, it's either only using the system clock as part of a calculation or not using it at all. The fact that our "successes" aren't always very close together suggests something else is causing the collision.

Also, we can have all the successes we've been having, but even having only one failure means it's not fixed - ie there's still a bug there we're just hitting it less now that we're specifically testing for it.

We could speculate to the cause of this error, but unless we see the source, or at least the message id calculation it will only be speculation.

So I'm thinking we just let gOOrOO know there is this bug, he will know if he did anything to address it in A32, and if so maybe it's partially fixed and were seeing more successes now than failures, or if he didn't do anything on this bug and we're just fluking it :)

Andrew
--- MagickaBBS v0.6alpha (Linux/x86_64)
* Origin: Cauldron BBS - Magicka Test BBS (21:1/125)

On 04/26/17, Black Panther said the following...

I've now completed 4 tests with Mystic utilizing MultiMail to send multiple messages in the same .REP packet. Each of them was successful,
as each of the messages contained a different MSGID. Here are the IDs
from the most recent test in fsxNet:

Wow! If I can use Multimail with Mystic, that will be great. I will
definitely love this. I am looking forward to using Multimail with Mystic. :)

--- Mystic BBS v1.12 A31 (Linux)
* Origin: A 90's Manila BBS telnet://a90smanilabbs.ddns.net (21:1/192)

On 04/27/17, Avon said the following...

Thanks Dan and Brian... well Agency is running on a Win 7 box that has a core 2 duo in it and 4 gig RAM. I don't think that's overly slow. I am
not sure the speed of the system is the contributing factor. What would
be good to try and figure out are the steps / software etc. taken by
each of you to get the results you got.

Brian testing on Dans system is a good idea, Dan perhaps you may want to try a few uploaded packets on Agency - also option O on the message menu (hidden option) .. I wonder if the issue is related to multiple posts in the same .REP packet that are replies to existing messages. I note that Dan created two 'fresh'/non reply posts and got them to generate with different IDs.

Hadn't thought of that, but would make some sense that replies are different from new posts. I had tried one reply and two new posts on Dan's system, but he's customized the message menu so no v to check MSGID so no idea if it
worked or not! I'll check the next few messages then try some fresh postings
to see if the MSGID works for new posts on Agency.

--- Mystic BBS v1.12 A31 (Raspberry Pi)
* Origin: Mystic Pi BBS bcw142.zapto.org (21:1/145)

apam wrote to bcw142 <=-

We could speculate to the cause of this error, but unless we see the source, or at least the message id calculation it will only be speculation.

So I'm thinking we just let gOOrOO know there is this bug, he will know
if he did anything to address it in A32, and if so maybe it's partially fixed and were seeing more successes now than failures, or if he didn't
do anything on this bug and we're just fluking it :)

Andrew
--- MagickaBBS v0.6alpha (Linux/x86_64)
* Origin: Cauldron BBS - Magicka Test BBS (21:1/125)

Yup, A32 is now out and should have all that fixed. Testing right now by using MultiMail v0.49 under the Pi to reply and putting a number of replies in one rep packet. We'll see, but it's been reported to be fixed and I have my Alpha Test system setup with it which I'll use to send this.
It seems it was something else, nothing to do with processor speed like I had guessed. Still I think g00r00 has now fixed all the little MSGID bugs now and A32 should be clear of them. This one is 64 bit Ubuntu Linux, I'll setup my main Pi based system later in the week (or weekend) and test that as well.
A32 is now up on the main site http://www.mysticbbs.com/ for all systems.


... MultiMail, the new multi-platform, multi-format offline reader!
___ MultiMail/Linux v0.49

--- Mystic BBS/QWK v1.12 A32 (Linux)
* Origin: Mystic AlphaTest bcw142.zapto.org:2323 (21:1/145.2)

On 04/26/17, Black Panther said the following...

I noticed you had stopped by to test it out. Sorry, I had the access set higher to avoid any problems. I have lowered the access to visiting sysops, so if you'd like to stop back and try it out. The option is "O" from the message menu. You should not be able to use it... :)

Yes if it might work better from a slower system, but might not be programmed that way (likely isn't). It's more likely it won't make any difference
coming from a slow system as it likely downloads, then processes. That would make the speed the same. Since it worked for you, it should work.

--- Mystic BBS v1.12 A31 (Windows)
* Origin: Mystic AlphaTest (21:1/145.2)