pan-users
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Pan-users] bug in 0.14?


From: Alberto BARSELLA
Subject: Re: [Pan-users] bug in 0.14?
Date: Sat, 10 May 2003 08:53:21 +0200
User-agent: Mutt/1.4.1i

Hi,

On Thu, 08 May 2003 at 03:21:38PM -0500, address@hidden wrote:

to school. When I came back, pan was loaded, the file was in queue and in the Task Queue it claimed to be dling at XXk/s with XX minutes left to download. However,[.....] so, is this a known bug, my news server, or both? And if its not a bug, what can I do about it.

Same here.  I have an unreliable ADSL connection.  I have a script
checking for a working connection and reestabilishing it in case it
drops, but the new version of pan is unable to correctly resume
downloads.  This is actually very annoying for me, since it means that I
cannot run pan unattended for a long time (i.e. one night).

The problem is with the new version of pan and not the newsserver.
My feeling is that the timeouts on the read from a socket are infinite
or almost infinite.  I'm having a look inside sockets.c, which seems to
be the place handling the "low-level" transfers across connections.
In the new version everything is done through glib/gnet, contrary to the
old version, and since I'm unfamiliar with those libraries it'll take
some time.....

If I find a fix, I'll post it.... but the problem is also that I've not
been completely able to reproduce the lockup systematically. If I
intentionally drop the connection, sometimes pan realizes this and
stops, restarting when the connection is reestabilished, sometimes it
just waits forever.

Bye,
Alberto
--
Alberto BARSELLA
** Beliefs are dangerous. Beliefs allow the mind to stop functioning.
A non-functioning mind is clinically dead.  Believe in nothing... **




reply via email to

[Prev in Thread] Current Thread [Next in Thread]