[Pan-users] Re: Very odd behaviour

[Duncan]

Bruce Bowler <address@hidden> posted
address@hidden, excerpted below, on  Thu, 06 Sep 2007
17:57:20 +0000:

> I've posted a couple of png files to http://www.bigelow.org/~bcb/
> prob1.png and prob2.png.

[I've had this saved unread for a couple days, as I was simply too tired 
to make sense of it.  I slept ~16 hours yesterday, which I had off, and 
it makes a lot more sense now! =8^) ]

OK, the files are there, but not directly reachable from the URL you 
posted.  That URL has a simple mailto: link, nothing else.

Next time, please either post the specific URLs, as below, or put the 
links in your index file served from the main URL given above.  I 
initially thought you had pulled them since they weren't visible or 
linked at the given URL. =8^(

This works:
http://www.bigelow.org/~bcb/prob1.png
http://www.bigelow.org/~bcb/prob2.png

> First, the text that gets shown has NOTHING what-so-ever to do with the
> item in the header pane and second, I can see no reason why these 2 got
> threaded together (other than the headers are "subject" and "Re:
> subject".
> 
> What's going on?

I see from your followup (which doesn't quote the problem pngs, thus my 
replying here) that you think it's the server.  I think so too.  There 
are two ways to ID an article, the article-id (in the xref header, 
generally sequential but server and group specific) and the message-id 
(globally universal ID, should /never/ be duplicated).  The server's 
article-id <-> message-id mapping appears to be messed up.

> pan 0.132, "Waxed in black"
> fc6

Thanks for including the version info.  A lot of folks forget that, and 
sometimes it makes a big difference.

If you get a useful reply from the query you sent them, please post it.  
I'd like to verify my guess and am curious to see what if anything they 
say about what might have caused the problem.

-- 
Duncan - List replies preferred.   No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master."  Richard Stallman