On (18 Jun 95) Rod Speed wrote to Keith Richardson...



 KR> my mail last thursday was .th2, followed on friday by .th3. dunno

 KR> what happened to .th0 and .th1. i certainly didn't do any previous

 KR> connects, and the .th2 was a normal size for a daily packet.



 PE> Ok, what happened was that on the 19/5, your system rejected

 PE> 0000fffa.th1, which I tried to send you.  This took it out of

 PE> my .HLO file, but left the th1 file sitting there.  Indeed, the

 PE> TH1 file is still sitting there doing nothing, and won't be sent

 PE> to you because it's not in the HLO file, and whenever squish needs

 PE> to create a new archive, it sees the TH1 sitting there and creates

 PE> a TH2 and then adds the TH2 to the .HLO file.



 KR> sounds a great theory paul, but what happened to .th0?



 RS> Presumably it had been successfully sent the day before the attempt

 RS> to send the .TH1 failed.



since the day before was wednesday, that would have been a major brain

fart, and since .we1 was received on the same mail run as .th2 it seems

exceedingly unlikely.



 KR> exactly the same thing happened this thursday, and i was

 KR> watching this time. nothing funny happened, just the receipt

 KR> of .we1, then .th2. next day .th3 arrived as before.



 RS> Thats just the old refused .TH1 still there on Pauls causing that.



so paul has already said, but, if there is a refused packet sitting

there on tml, why aren't i missing any mail, and why was ,th2 the

normal size for a daily mail packet?



 KR> unfortunately the bink log gets overwritten each day, but

 KR> i will save it in future. i see no reason that, after 12 months

 KR> with no changes, bink would suddenly start to refuse mail.



 RS> Sure, but then computings like that. When you do eventually find out

 RS> why it has suddenly started to do something different, you usually can

 RS> see why it did.



yes rod, in the 25 years that i've been making a living out of

computers, i have come across such things. it is indeed possible that

bink 2.56 has such a bug, although, given that it has been out now for

2.5 years or so, it is relatively unlikely to have gone unreported. i

may get tempted to use paul's fudged version of bink as then i could

blame him no matter what happens, but i am reluctant to give up a stable

reliable product for a potential can of worms.



 KR> the system here is being rebuilt from end to

 KR> end to integrate the new 2.1 gig 7200rpm drive,



 RS> Funny that, and you are surprised it does something different

 RS> to what it has been doing for the last 12 months ?



yep.



 RS> Thats absolutely classic, if things start to be seen to be

 RS> different, and you have just changed your system, very very likely

 RS> indeed its not a coincidence. Tho sometimes it is anyway.



having two bob each way here? a singular event some two weeks after the

change seems unlikely to be related, especially since the entire

directory structure was simply copied to another partition, and then

back with the only difference being that it is now on a c: partition of

320 megs instead of a c: partition of 160megs.



 KR> but the only thing that has happened to the point

 KR> is to copy it to another partition, and then back

 KR> to it's own, it is even on the same physical drive.



 RS> Heard it all before Keith. Time will tell what its due to, but I wont be

 RS> exactly shocked speechless if it does turn out to be due to your reconfig.



we'll see, but with 11 mail runs since and no re-occurrence the

probability of it being a singular event rises every day, and if it

proves to be such then the likelihood of it being caused by the

re-configuration is 3 parts of 1% of absolutely bugger all (:



                        Keith



... God is love. Love is blind. Therefore, Ray Charles is God.



--- PPoint 1.88