>Well, that was a known bug in 1.93 which was fixed in 1.94. If something
>else is causing lists to go away, I haven't heard about it.
You have now. I've been using 1.94 since it came out. One of my lists
has been wiped twice, reduced from almost 400 subscribers to zero.
Also perl 5.003, SKI PI 4D/25 running IRIX 4.0.5.
At least one case occurred when I submitted several messages for
approved unsubscriptions close to each other, so one presumably arrived
before the last one completed processing.
>If (as in the 1.94 distribution) we don't know that any problem exists, why
>should we warn about it? "We regret to inform you that we don't know of
>any problems, so to avoid them, be sure to..."
Fine, now that you know the problem exists, I suggest that the
documentation be adjusted while we try to figure out what went wrong.
>Honestly, have you seen a single complaint of list erasure with 1.94? I
>haven't, and I've tested it under all of the conditions which would kill
>1.93. I cannot reproduce the problem. I've driven my load above 500
>sending ten thousand subscription requests, during which I filled up the
>partition. The list was not erased (though of course some requests were
>lost, but if you can't append to the list, you just can't).
As written above, I have. And keep in mind that I am not an expert on
this, and it may well be that I have some permission wrong (although I
have run the test suite) or committed some other blunder that you have
not. Still, there is a problem and I'm willing to work on fixing it, but
I made some recommendations that would make Majordomo more bulletproof.
Bill Silvert, Habitat Ecology Section, Bedford Institute of Oceanography,
P. O. Box 1006, Dartmouth, Nova Scotia, CANADA B2Y 4A2, Tel. (902)426-1577