• fe 1.46

    From Bill Gordon@1:123/22 to All on Wed Jul 21 13:57:18 2004
    I'm a registered owner of FE 1.45, and it works great. My question is this: I've heard horror stories about v1.46 and want the true skinney. Is it stable,
    is there a fix for the problems with it, or are there really any problems.

    Thanks....


    * Origin: *Square One telnet://squareone.no-ip.org (1:123/22)
  • From Renato Zambon@4:801/161 to Bill Gordon on Thu Jul 22 10:36:58 2004
    I'm a registered owner of FE 1.45, and it works great. My question
    is this: I've heard horror stories about v1.46 and want the true
    skinney. Is it stable, is there a fix for the problems with it,
    or are there really any problems.

    Using FE 1.46.1 (first DOS and now DPMI flavour) here ok for years
    with RA, JAMbase for all areas, with currently 27 links and +1200
    areas.

    []s

    ---
    * Origin: HidraSoft BBS * SP,Brasil * telnet hidrasoft.dyndns.org (4:801/161)
  • From mark lewis@1:3634/12 to Bill Gordon on Thu Jul 22 13:25:28 2004
    I'm a registered owner of FE 1.45, and it works great. My
    question is this: I've heard horror stories about v1.46 and
    want the true skinney. Is it stable, is there a fix for the
    problems with it, or are there really any problems.

    i don't know what supposed horror stories you've heard but i'm not aware of any... i've run every version since FE was released and have never had any problems that weren't self induced... v1.46 is very stable and there was only one fix needed for it and that only if you run in a binkley type setup...

    )\/(ark


    * Origin: (1:3634/12)
  • From Bill Gordon@1:123/22 to mark lewis on Fri Jul 23 00:39:58 2004
    i don't know what supposed horror stories you've heard but i'm not aware
    of any... i've run every version since FE was released and have never had any problems that weren't self induced... v1.46 is very stable and there
    was only one fix needed for it and that only if you run in a binkley type setup...

    )\/(ark

    Thanks, Mark, I think the horror stories I heard were, indeed, about binkley type mailers. I was running binkD, but finally figured how to configure Radius, another binkley type mailer. I have been using 1.45 for so long that I
    wonder if it's worth moving to 1.46.


    * Origin: *Square One telnet://squareone.no-ip.org (1:123/22)
  • From Paul Quinn@3:640/384 to Bill Gordon on Mon Jul 26 23:04:00 2004
    Hi! Bill,

    In a message to Mark Lewis you wrote:

    I have been using 1.45 for so long that I wonder if it's worth
    moving to 1.46.

    The only true bug I've seen with 1.46 is in using its FeSetup.Exe; I found that
    it just would not accept the "[" & "]" chars when describing directory paths which can be extracted from DOS environment variables. Keep your FeSetup.Exe from the 1.45 release, at least. I use DOS envars extensively and had to FREQ a 1.45 archive from a feller in Canada (I think) via the POTS to fix my config at some particular time.

    Cheers,
    Paul.

    --- Radius 4.010-2004/04/21_Beta
    * Origin: Computer... smoke... uh oh... (IP 1200-1500 UTC) (3:640/384)
  • From Bill Gordon@1:123/22 to Paul Quinn on Mon Jul 26 11:37:26 2004
    The only true bug I've seen with 1.46 is in using its FeSetup.Exe; I
    found that it just would not accept the "[" & "]" chars when describing directory paths which can be extracted from DOS environment variables.

    Well, can't those characters be changed??

    Keep your FeSetup.Exe from the 1.45 release, at least. I use DOS envars extensively and had to FREQ a 1.45 archive from a feller in Canada (I
    think) via the POTS to fix my config at some particular time.

    I have the archives, both the v1.46 AND the v1.46x, but even with ANY DMPI versions, all I get it trouble. Seems my system does not to run anything in DMPI mode, either allfix, amu, or fastecho 1.45.

    Maybe I'd better leave well enough alone, huh??

    Thanks...


    * Origin: *Square One telnet://squareone.no-ip.org (1:123/22)
  • From Paul Quinn@3:640/384 to Bill Gordon on Tue Jul 27 16:48:00 2004
    Hi! Bill,

    On Mon, 26 Jul 04, you wrote to me:

    I found that it just would not accept the "[" & "]" chars when
    describing directory paths which can be extracted from DOS
    environment variables.
    Well, can't those characters be changed??

    Probably. But I haven't bothered with hex editors since utilizing the autoexec
    function on CP/M 2.xx diskettes... ahmm, mid-80's. :)

    I have the archives, both the v1.46 AND the v1.46x, but even with ANY
    DMPI versions, all I get it trouble. Seems my system does not to run anything in DMPI mode, either allfix, amu, or fastecho 1.45.
    Maybe I'd better leave well enough alone, huh??

    Damned right. So what! if the mail takes a few extra milliseconds to toss these days.[giggle] There's more to life than fixing what ain't really broke.

    Cheers,
    Paul.

    --- Radius 4.010-2004/04/21_Beta
    * Origin: Dammit! I wanna be _their_ aliens. (IP 1200-1500 UTC) (3:640/384)
  • From mark lewis@1:3634/12 to Bill Gordon on Mon Jul 26 21:28:58 2004
    i don't know what supposed horror stories you've heard
    but i'm not aware of any... i've run every version since
    FE was released and have never had any problems that
    weren't self induced... v1.46 is very stable and there
    was only one fix needed for it and that only if you run
    in a binkley type setup...

    Thanks, Mark, I think the horror stories I heard were, indeed,
    about binkley type mailers.

    if they were concerning v1.46, they should have been shortlived... one thing that i do know is that the BSO format can be problematic if one doesn't take into account what their default zone is and also doesn't take into account that
    zone numbers are represented in HEX...

    I was running binkD, but finally figured how to configure
    Radius, another binkley type mailer.

    yes, i run it here on one of my test/point systems but not with FE (yet)...

    I have been using 1.45 for so long that I wonder if it's worth
    moving to 1.46.

    since this version has been out as long as it has, i'd say that it is quite safe to update to it... you shouldn't have any problems once everything is in place and updated properly... i went thru several upgrade routines and have never had any problems...

    i don't see/remember anything like a changelog but here's the whatsnew file for
    v1.46pl1 (that's pee ell one, the bink bug patch)...

    Changes since FastEcho 1.45a
    ============================

    Version 1.46pl1, 03 Apr 1997
    ----------------------------
    Small patch #1

    * FastEcho PACK -P for Binkley-style outbounds was ignoring "priority"
    NetMail messages which had ^aFLAGS like DIR, IMM and others unless
    they were InTransit, fixed


    Version 1.46, 01 Apr 1997
    -------------------------
    Late Easter Update

    * including OS/2 and DOS/32 versions of FESetup using the Watcom C++
    compiler, DOS/32 FastEcho and FEUTIL versions (replacing the /16
    ones) can be expected once they got ported to Watcom as well

    * added support for 6 more groups 1-6 and a maximum of 4096 areas

    * AreaFix is now able to send the rules of newly linked areas to the
    requesting system based on *.RUL files with a definable "Rule file
    area tag prefix" for .RUL files not following the current standard
    supported by FastEcho (as used at least in Germany). FastEcho will
    parse the files for a line starting with that prefix being followed
    by the name of the area

    * added "delete area" (~AREA) and "rename area" (#OLD NEW) commands to
    AreaFix, changes can be automatically forwarded to your downlink's
    AreaFix (or as a human readable information message)

    * you can now disable the forwarding of AreaFix requests for single
    systems, see Node Manager for the new "AreaFix flags" settings

    * added a deny list to the forward AreaFix section which includes
    names (wildcards allowed) of areas which may never be forwarded even
    if found in the specified list

    * added option to tell AreaFix not to set forwarded areas to
    Passthrough and rather use the Storage method defined in the
    corresponding Group Defaults

    * added FastEcho RELINK command to send AreaFix requests to selected
    systems or all your uplinks requesting all areas for which the
    systems are active

    * now creating a FEPACK.REC file for recovery when FastEcho is inter-
    rupted while archiving ARCmail (hang), on the next run it will look
    for this file and rename the .PKTs found in there back to .QQQ to
    process them again, also cleans up the mailer's busy semaphore file
    plus a temporary list file (usually FETMP*.*) found in there,
    FEPACK.REC is stored either in a directory SET FETMP= is pointing
    to, or in your TMP/TEMP or the current directory (due to the nature
    of that file it's suggested not to place it on a ramdisk)

    * expanded ARCmail pack priority from Y/N to a 4-level setting, 0
    being the lowest, 3 the highest, so ARCmail for systems with a level
    of 3 will be compressed first, then level 2 and so on

    * added "Minimum packet size" setting to specify a value below which
    inbound EchoMail packets will be skipped (check the Help text for
    more details) with FEOPT=NOLIMIT or TOSS /NOLIMIT to disable the
    checking of the "Minumum packet size" for the current run. Packets
    containing NetMail will be tossed in any case

    * added error reporting via NetMail (can be disabled with SET FEOPT=
    NOERREP)

    * now supports stripping of SEEN-BY info for outbound out-of-zone
    EchoMail messages (additionally to the stripping when acting as an
    inbound zonegate), by default disabled (because it's slower) can be
    enabled with SET FEOPT=ZONEGATE

    * added -S switch to FastEcho STAT command to suppress the system's
    statistics thus create an area-only statistics

    * AreaFix is now able to understand the '=area' command correctly,
    which makes it able to perform a rescan for all linked areas
    with '=*' without adding any new areas

    * added 2 new AreaFix meta commands, %EXPORT and %NOEXPORT, to turn
    export by name on/off (mostly used in conjunction with %PASSIVE)

    * now allowing local AreaFix requests written locally with a FROM:
    line in them, you have to add a system using the used AKA to the Node
    configuration

    * AreaFix will no longer include a list of all linked areas in the
    receipt message if the request message has the PGM (program
    generated) ^aFLAGS

    * AreaFix now stops processing a request when it encounters %QUIT or
    "-- " (as internet signature start)

    * When %AVAIL is requested and no forward lists are entered in FESetup,
    AreaFix will respond with a receipt showing a proper error message

    * FastEcho TOSS: added new setting 'H' = Hide for Keep Seen-By (in
    addition to 'Y' and 'N') to have the seen-by info hidden as a kludge

    * FastEcho TOSS: the dupe checker is now ignoring short (and thus
    invalid and most likely duplicate) MSGIDs (like "^aMSGID: fae"),
    only MSGIDs longer than 14 characters will be used for dupe checking

    * FastEcho TOSS -C does no longer make carbon copies of messages tossed
    to your BadMail area

    * FastEcho TOSS: changed the code handling the check for used space on
    the temporary outbound drive slightly

    * FASTECHO TOSS is no longer stripping DIR/IMM FLAGS from incoming
    NetMail messages

    * FEOPT NOPACK is no longer honored if FastEcho TOSS is forced to
    compress the outbound .QQQs when one of the preset limits is reached

    * FastEcho SCAN: now resolving the paths in ECHOMAIL.JAM when comparing
    them to the paths of the areas (truename function), can be turned off
    with SET FEOPT=NORES or SCAN /NORES

    * FastEcho SCAN: now only logging ONCE if a message in a JAM/Squish/
    .MSG area couldn't be exported, instead of doing that for every
    message

    * FastEcho SCAN JAM NetMail export now properly supports the way JAM
    can store File attaches and requests

    * added ^aFLAGS NPD and PGM for JAM EchoMail in TOSS and SCAN and for
    JAM NetMail export, same for FEUTIL POST (to all messagebase types)
    as well as Netmail messages created by FastEcho

    * FastEcho PACK does now ignore "unknown" ^aFLAGS (instead of not
    packing NetMail messages with these flags as earlier versions),
    added support for the DNP (Do Not Pack) flag, also NetMail messages
    having the HUB flag will not be packed unless they are InTransit

    * FastEcho PACK -P does now properly handle FileRequests with password
    (subject "file !pw")

    * when packing priority NetMail for Binkley-style mailers, FastEcho
    PACK should now handle otherdomain outbound directories better for
    out-of-own-zone mail (ie. it will use the domain defined for the
    AKA which matches the sender's zone, you probably will have to tell
    your mailer which zones that are not covered by one of your AKAs
    belong to which domain)

    * FASTECHO PACK will now pack NetMail messages which have the IMM/DIR/
    CRA/HLD/FRQ FLAGS set but are InTransit (non-local), same for
    messages with KFS/TFS flags but without a FileAttach flag

    * automatic passive for systems didn't work if an ARCmail address
    different from the main address was used and no other system was
    listed having this address as main address

    * automatic passive will now set all systems which have the same
    ARCmail address as the new passive one to passive as well

    * you can now use $[COMSPEC] (or any other macro instead of the program
    name) for any external programs, previously no macro expansion was
    supported

    * FastEcho will now also look for other executables (like the packer)
    in the directory where .EXE is located

    * the 'list' file for packer calling is now created in the TEMP or TMP
    directories (if SET)

    * FDRESCAN.NOW is now created in every *.MSG directory in which
    changes are performed (only if your mailer is Frontdoor, of course)

    * fixed a bug in the ROUTE.FE parser which assumed a wildcard for all
    points when you do a "Direct z:yyy/xxx" (which was translated to
    z:yyy/xxx.* instead of z:yyy/xxx.0)

    * fixed a little problem with very large arcmail bombs where the pack
    ratio wasn't calculated correctly

    * fixed FastEcho SCAN calculating a different dupe CRC than the TOSS
    routine, thus causing messages, which were written locally not being
    detected as dupes by FastEcho TOSS

    * fixed a problem with AreaFix %LIST not properly formatting some
    comments

    * fixed Carbon Copies to JAM Echomail areas not getting the Local flag
    and thus not being exported by SCAN

    * fixed some LZH archive parsing problems

    * fixed auto-area-create not expanding macros for the messagebase
    directories found in the Group defaults

    * FastEcho TOSS did not always return proper error codes (such as
    when the HMB was too big), fixed

    * FastEcho was opening the .CFG in DENYALL instead of DENYWRITE mode,
    fixed

    * now supporting /? additionally to ? on the command line to show the
    usage screens

    * added -G<groups> switch to FESETUP -XA, -XE and -X$ to have only
    areas belonging to certain groups exported

    * added support for GEcho 1.20 and FMail 1.22 import to FESetup

    * FESetup: when creating a new FASTECHO.CFG the De/Compression program
    defaults will be copied copied without having to enter the De/Compr.
    programs menus

    * FESetup was not resetting the 'Passive' flag when a passthrough area
    was changed to a different storage type, thus these areas (even
    though not really passive) were not included in the AreaFix lists

    * speeded up FEUTIL2 JAM LINK quite a lot

    * FEUTIL IMPORT does no longer store FMTP/TOPT/INTL as JAM subfields

    * FEUTIL JAM POST is now correctly updating ECHOMAIL/NETMAIL.JAM

    * FEUTIL POST did not expand the macros in the paths of FIDO and JAM
    areas, fixed

    * lots of other smaller fixes and cosmetical changes




    )\/(ark


    * Origin: (1:3634/12)
  • From mark lewis@1:3634/12 to Paul Quinn on Mon Jul 26 21:33:16 2004
    I have been using 1.45 for so long that I wonder if it's worth
    moving to 1.46.

    The only true bug I've seen with 1.46 is in using its
    FeSetup.Exe; I found that it just would not accept the "[" &
    "]" chars when describing directory paths which can be
    extracted from DOS environment variables. Keep your
    FeSetup.Exe from the 1.45 release, at least. I use DOS envars
    extensively and had to FREQ a 1.45 archive from a feller in
    Canada (I think) via the POTS to fix my config at some
    particular time.

    oh yes! there is/was that bug... i resorted to hexediting the config file IIRC when i was using/testing those... i had forgotten all about that one... in my case, i did what i found obvious... i put !'s for the [ and ] and since it was only a couple of paths that i was doing this on, i only had one or two places to hexedit...

    i guess it should be mentioned that one should be careful when using an older FESETUP due to new capabilities/settings that may get overwritten/lost with new
    config files...

    )\/(ark


    * Origin: (1:3634/12)
  • From mark lewis@1:3634/12 to Bill Gordon on Mon Jul 26 21:36:30 2004
    The only true bug I've seen with 1.46 is in using
    its FeSetup.Exe; I found that it just would not
    accept the "[" & "]" chars when describing
    directory paths which can be extracted from DOS
    environment variables.

    Well, can't those characters be changed??

    the problem is that FE validates the characters that can be used in a directory
    and file name...

    Keep your FeSetup.Exe from the 1.45 release, at least.
    I use DOS envars extensively and had to FREQ a 1.45
    archive from a feller in Canada (I think) via the POTS
    to fix my config at some particular time.

    I have the archives, both the v1.46 AND the v1.46x, but
    even with ANY DMPI versions, all I get it trouble.
    Seems my system does not to run anything in DMPI mode,
    either allfix, amu, or fastecho 1.45.

    what OS are you running? what memory managers and such?

    Maybe I'd better leave well enough alone, huh??

    you don't have to run the DPMI version(s) of anything unless you want to and/or
    want to take advantage of some feature(s) they might offer... in the case of FE, larger messages is one benefit...

    )\/(ark


    * Origin: (1:3634/12)
  • From Paul Quinn@3:640/384 to Mark Lewis on Tue Jul 27 22:47:00 2004
    Hi! mark,

    On Mon, 26 Jul 04, you wrote to me:

    FeSetup.Exe; I found that it just would not accept the "[" &
    "]" chars when describing directory paths
    [ ...trim... ]
    i guess it should be mentioned that one should be careful when using
    an older FESETUP due to new capabilities/settings that may get overwritten/lost with new config files...

    In this case there were no other ill effects. Lucky me. The edits went smoothly and the 1.45 .Exe was discarded immediately afterwards.

    (Of course, I did take the precaution of having several copies of the previous FastEcho.Cfg available, if the whole exercise went 'belly up'. ;-)

    Cheers,
    Paul.

    --- Radius 4.010-2004/04/21_Beta
    * Origin: It works better if you plug it in. (IP 1200-1500 UTC) (3:640/384)
  • From Bill Gordon@1:123/22 to Paul Quinn on Tue Jul 27 12:48:44 2004
    Probably. But I haven't bothered with hex editors since utilizing the autoexec function on CP/M 2.xx diskettes... ahmm, mid-80's. :)
    Those were the days. Things were so much simpler then. I think, anyway...

    Damned right. So what! if the mail takes a few extra milliseconds to
    toss these days.[giggle] There's more to life than fixing what ain't
    really broke.
    Well, FE works fine for me, I really don't have any callers, so to speak, and what little echoes I subscribe to toss just fine.


    * Origin: *Square One telnet://squareone.no-ip.org (1:123/22)