Hi,
 
This is my first post.
 
Maybe someone can explain this to me.
 
I'm using curl 7.13.2 (i386-pc-win32) with libcurl/7.13.2 OpenSSL/0.9.7e
zlib/1.2.1

I have a script that tries to download a daily file called ctaschq.dat,
also a monthly file called AccountExtractYYYY-MM-DD.csv,
where YYYY-MM-DD would be the current date. The monthly file is produced
on the last business day of the month, or the next business day of
following month if it's a weekend or holiday.
 
It's also possible that there may be no files on a particular day or
multiple files.
 
If there are multiple files they are named like:
ctaschq_N.dat or AccountExtractYYYY-MM-DD_N.csv, where N = {1, 2, 3
...}
 
The way the script works, is it tries to download the file, say
ctaschq.dat, and if there's been a download, a counter is set, and the
script attempts to download ctaschq_1.dat, and so on. This continues
until the error code returned is not 0.
 
When it's not zero, it checks the error number and writes the message to
the event log and also to an HP Openview monitoring application.
 
Since a file may or may not be present, I don't consider error code 22
(not found) an error, and don't display it on Openview.
 
What I found today when this ran (the first time there was a monthly
file) was that the error code 22 (not found) was returned for
ctaschq_1.dat, but for AccountExtract2005-05-31_1.csv, I get error code
28 (timeout).
 
The problem with this error code 28 is that I consider a timeout an
error condition and report it as such.
 
I see no reason why it would return this error code instead of 22.
There's definitely no such file on the remote server, as I logged in
manually and checked.
 
Can anyone shed any light on this?
 
 
 
Received on 2005-06-01