Martin Schulze wrote:
> Now what happens is, hypermail tries to write that new file as 0210.html
> instead of 0346.html. Why? Wouldn't it be more expectable that the
> file 0210.html would be ignored and a new file would be created?
> Especially since writing to 0210.html fails?
I think this happens because it failed to read the next file (in this case, 0210). It successfully found out that file number 0209 exists, and suddenly encounters a problem at the next one and stops. And for some reason, assumes that's the next one it should be writing.
Personally, I think this is a problem, yes. A problem in the sense that hypermail stops and tries to write that file out instead of stopping with an error. But at the same time, how to address it? I mean, you're talking about one single file that got chmodded. What if someone else decides to chmod a whole slew of files? Is hypermail supposed to ignore each error it encounters and just continue increasing its counter? Maybe a better way was if hypermail kept some sort of record of the last file it wrote (0346) and on the next run just increment from there? This smells like a DB capability (can anyone say mhonarc?)
AMK4
-- H | Hi, I'm currently out of my mind. Please leave a message. BEEEEP! |____________________________________________________________________ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Ashley M. Kirchner <mailto:ashley_at_pcraft.com> . 303.442.6410 x130 Director of Internet Operations / SysAdmin . 800.441.3873 x130 Photo Craft Laboratories, Inc. . eFax 248.671.0909 http://www.pcraft.com . 3550 Arapahoe Ave #6 .................. . . . . Boulder, CO 80303, USAReceived on Sun 17 Sep 2000 08:01:42 PM GMT
This archive was generated by hypermail 2.2.0 : Thu 22 Feb 2007 07:33:52 PM GMT GMT