• Date/Time Changes possible from Mystic - Causing Network dups

    From Allen Prunty@21:1/148 to Nicholas Boel on Thursday, November 03, 2016 12:33:00
    ** COPIED FROM FIDO_SYSOP

    Hello Nicholas!

    03 Nov 16 09:18, you wrote to me:

    However if Mystic was indeed actually doing this, we would be seeing it
    a LOT more often and from a LOT more systems (instead of /only one/), wouldn't we?

    Well if we take a trip to the theoretical hardware store and look at the hypothetical chainsaw we can deduce this.

    1. If the bug is the same on all mystic systems and the alteration is
    the same across all mystic systems then there would be TWO sets of
    duplicate timestamp changes with the same MSG Introduced into the
    fidonet ecosystem.

    2. The message in FIDO_SYSOP with the same msgid and timestamp of 12:05:36
    would be munged to 12:05:00 with the same MSGID. Basically, there are
    at this time TWO distinct dups. So basically the first person who
    mangles the timestamp with mystic would be the person who's message will
    judge all other messages that have duplication.

    3. Since Ben is the coordinator of the echolist he's pretty much connected
    to the top level hubs of Z1. So... with that he has the greatest chance
    of altering the message header and passing it on. Chances are he's
    sending out the 12:05:00 message that's cancelling all other mystic
    maniplulations that come after him.

    Ben's system is a major link in the Z1 Ecosystem... it's entirely possible it is him. I've also looked at the QWK alterations and checked the message on Ozzmosis... there was no configuration switch changed for the QWK import
    date. I called Mr. Coyle out on changing the dates to the import time and he said this is how he designed mystic. I then told him that I did not think
    this was the best way to do it as my users who were using a QWK network were complainging all the messzges had the same date. Furthermore the network participants on other systems were getting frustrated becuase the users were responding to messages that were years old.

    He then saw it my way (after a bit of a heated debate) and changed it to
    import to the date and time of the origional message. The problem is I no longer import the QWK network so I'm not sure how it's imported... the
    million dollar question is does it change all of them to :00 on the seconds
    or is it actually importing the seconds correctly.

    Which upon looking at my last import on the MAC that runs mystic... it
    appears that mystic by design is dropping the seconds even on the QWK.

    From DoveNet General / A QWK Network

    Mystic Header
    -------------
    From : Poindexter Fortran Msg # : 4221 of 4227
    To : Mro Msg Date : 10/28/16 08:15
    Subj : Fake moon landings, n Refer to : 0
    Stat : Echo Sent See Also : 0

    GOLDED JAM Header
    -----------------
    Msg : 4221 of 4226 Snt
    From : Mro 1:2320/100 28 Oct 16 12:29:00
    To : Poindexter Fortran 29 Oct 16 00:56:20
    Subj : Fake moon landings, n

    ==========================================================================

    For brevity, I'm not going to post any more headers than this because they
    all are indicating the same thing. Both in Fidonet and QWK upon my examination. The FROM dates on both the fido and the QWK import have the seconds changed to 00... consistently across the board. But not the TO.
    This seems like there is a bug on the routine that inserts the message into
    the Jambase. This is surprising since MUTIL imports Fido and QWKPOLL imports QWK. Either way I do maintain my position that either fido or QWK the date/time should NOT be altered under any circumstances.

    I'm going to cross post this in FSX_NET's Mystic conference so our Awesome friend Paul can do further verification on this.

    Allen


    ... A program is used to turn data into error messages.

    --- WILDMAIL!/WC v4.20a
    * Origin: LiveWire BBS -=* LiveWireBBS.com *=- International (21:1/148.0)
  • From Avon@21:1/101 to Allen Prunty on Friday, November 04, 2016 12:42:00

    On 11/03/16, Allen Prunty pondered and said...

    examination. The FROM dates on both the fido and the QWK import have the seconds changed to 00... consistently across the board. But not the TO. This seems like there is a bug on the routine that inserts the message into the Jambase. This is surprising since MUTIL imports Fido and
    QWKPOLL imports QWK. Either way I do maintain my position that either fido or QWK the date/time should NOT be altered under any circumstances.

    I'm going to cross post this in FSX_NET's Mystic conference so our
    Awesome friend Paul can do further verification on this.

    Hi Allen... I did post about this in Fido and confirmed what I thought was
    this behaviour happening also. Have a look at my posts in FN_SYSOP. I also cross posted my reply here when I did report my findings..

    So not sure what is 'new' about any of your conversation?

    Best, Paul

    --- Mystic BBS v1.12 A31 (Windows)
    * Origin: Agency BBS | telnet://agency.bbs.geek.nz (21:1/101)
  • From Paul Hayton@21:1/101 to Allen Prunty on Friday, November 04, 2016 16:33:00
    On 11/03/16, Allen Prunty pondered and said...

    Hi Allen... I did post about this in Fido and confirmed what I though was this behaviour happening also. Have a look at my posts in FN_SYSO I also cross posted my reply here when I did report my findings..

    So not sure what is 'new' about any of your conversation?

    The -new- part is that when it imports QWK net messages it is also
    zeroing the seconds on incoming messages in the FROM field... however,
    it does not zero the seconds on the TO portion of the message.

    Since QWKPOLL imports the QWK net messages and not MUTIL then QWKPOLL
    has the same bug.

    Ah OK thanks for this, I had not picked up those comments in the dialogue.
    I'll cross post this to fsxNet reply for completeness and it could be added
    to a list of things to let James know of / look in to when he's back in the coding/Mystic dev zone :)

    Best, Paul

    --- Mystic BBS v1.12 A31 (Windows)
    * Origin: Agency BBS | telnet://agency.bbs.geek.nz (21:1/101)
  • From Allen Prunty@21:1/148 to Paul Hayton on Saturday, November 05, 2016 00:52:00

    Hello Paul!

    04 Nov 16 16:32, you wrote to me:

    Ah OK thanks for this, I had not picked up those comments in the
    dialogue. I'll cross post this to fsxNet reply for completeness and it could be added to a list of things to let James know of / look in to when he's back in the coding/Mystic dev zone :)

    If you have been keeping a list can you please post it so we are all on the same page with the issues. This way we won't keep rehashing the same stuff.


    Allen


    ... Buffet: French for making a pig of yourself.

    --- WILDMAIL!/WC v4.20a
    * Origin: LiveWire BBS -=* LiveWireBBS.com *=- International (21:1/148.0)