pan-users
[Top][All Lists]
Advanced

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

Re: [Pan-users] Re: Question around multi-server and connection errors


From: Guilhem Bonnefille
Subject: Re: [Pan-users] Re: Question around multi-server and connection errors
Date: Tue, 24 Oct 2006 13:29:35 +0200

Interesting to know that other people seems to have the same problem.
A big info for me is to learn that pan "should" handle this.

This morning, I noticed that, after exactly one hour, without any
action made by me (humh, I think not), the 4 erroneous servers have
been connected once again.

Actually, I really don't have any time to investigate, but perhaps in
the future.

On 10/24/06, Duncan <address@hidden> wrote:
"Guilhem Bonnefille"
<address@hidden> posted
address@hidden, excerpted
below, on  Tue, 24 Oct 2006 08:50:45 +0200:

> I'm running the pan 0.117+patches. Group list can now be downloaded.
> Thanks again!

=8^)

> But I encounter a strange behaviour and I wish to have some explanations.
>
> I'm using 5 servers.Few hours before, I started todownload 50 messages
> (binaries one). Pan started to connect to the servers. For any
> reasons, it failed. Only one server replied, and a download started at
> 0.1KB/s. I just leave everything and went to sleep.
> This morning, I checked pan: no file downloaded and a transfert is
> still here at 0.1KB/s. So I decided to stop pan and restart it. Then,
> it contact all servers and the download start at 100KB/s (nice :-)).
> But, few minutes later, only one server is still connected and the
> download is around 40KB/s (nice too). In the log, 4 errors around
> socket.

For error prone server connections, pan continues to have serious known
problems.  This is one of them.  At this point, it's looking doubtful that
they'll be appropriately fixed before 1.0, tho as long as 1.0 isn't out
yet...

Anyway, what's apparently missing at this point is someone who can repeat
the issue reliably and has the dedication to work closely with Charles,
debugging and troubleshooting until it's fixed.  Someone that can do C++
would be a bonus, but shouldn't be absolutely required.  However, time and
patience likely will be, as from the reports, there may be more than one
bug in this area and it may take several rounds of troubleshooting to pin
things down.

I'm not actually sure of the status of any related bugs on this at this
point. Last I looked, everything open seemed to be marked enhancement,
which would seem to indicate there wasn't a bug open on this altho it has
been reported in the groups.  If you believe you have the time and
dedication necessary for the possibly several rounds of troubleshooting
and debugging sessions, I'd suggest double-checking that one hasn't been
filed in the last few days and filing one if not.  It's known as it has
been reported in the groups, but without someone who can duplicate the
problem willing to file a bug and do the necessary troubleshooting with
Charles, the problem remains.

Fortunately for me but unfortunately for fixing this problem, my
connection seems to be trouble-free enough to trigger something like this
only rarely, which means it's not something I could effectively
troubleshoot.

--
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



_______________________________________________
Pan-users mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/pan-users



--
Guilhem BONNEFILLE
-=- #UIN: 15146515 JID: address@hidden MSN: address@hidden
-=- mailto:address@hidden
-=- http://nathguil.free.fr/




reply via email to

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