I do not (and do not wish) have a login on the bug database. Could you open the bug for me, please?
I joined the --debug log file.
Thanks in advance.
On 10/16/06,
Duncan <
[EMAIL PROTECTED]> wrote:
"Guilhem Bonnefille"
<[EMAIL PROTECTED]> posted
[EMAIL PROTECTED] , excerpted
below, on Sun, 15 Oct 2006 23:10:10 +0200:
> I'm quite surprised with this new version. I set some servers before
> refreshing group list but I'm unable to collect group list. After few
> hours of "refreshing" one of my pan instance retrieves 440000 groups, on
> each of my desired servers ( news.free.fr and allnews.readfreenews.net).
> I'm not sure these servers have such list. I think they have something
> like 120000 groups. It seems that something goes wrong here, like a deadly
> loop.
I suspect that's an already known bug introduced to 0.116. It was related
to clearing a net input buffer when it should have been keeping the
existing content and continuing to add to it. This was obviously a
show-stopper for 1.0, but the fix will be in 0.117 (which with luck will
be the final beta before 1.0), or it's in CVS (what I'm now running)
already. See the list of bugs already addressed for 0.117, based on the
CVS version, that I posted about 8 hours ago (new thread, mentioning the
new icons) for the bug number along with those for other changes. There
wasn't a patch on the bug but a line number and change, so you can patch
it manually if you wish to compile from tarball but don't want to do CVS.
An alternative would be rolling back to 0.115 temporarily, as it was
slower, but didn't have this corruption. Packages for that should still be
available if you aren't interested in compiling.
> Then, on an other user, I started the new pan, but I act differently. Now,
> I only ran pan some minutes. So it collected only 30000 groups info on
> each servers. Then I closed it and restart it. The group list is always
> empty. Where are my 30000 groups?
pan doesn't always write stuff to disk immediately. I'm guessing since it
aborted the download, it didn't write the list out to disk, so it's not
there to be read back in. However, getting that erasing buffer thing
fixed should eliminate the problem and allow you to download the entire
list without taking forever, so this shouldn't be an issue as soon as that
part of pan's working correctly.
> I also noticed that, each time I consult the servers list, when I close
> this dialog, all the currently active tasks are cloned: I started with 2
> group list refreshing tasks, I opened and closed the server list dialog
> and then I have 4 tasks.
I don't believe I've seen this bug reported yet. I'm guessing most folks
set their list and then don't bother it, so never come across the bug. As
I said, 1.0 is imminent, so please post this bug ASAP so it can get fixed
before 1.0. Once you do so, please post the bug URL or number here, and
I'll test and either confirm it or say it's /not/ happening here, giving
Charles a second point of reference. Again, please bug it right away or
it may not be picked up in time for 1.0. (Charles will probably see it
here but doesn't follow this as closely as he does bugs, and might miss
it, too.)
> Any help would be appreciate.
>
> PS: I'm running Debian GNU/Linux in testing flavour, and some "unstable"
> packqges (like pan).
=8^) Gentoo (amd64, dual Opteron, so you see why compiling is no big deal
for me) here. However, the Debian maintainer (Soren B Overgaard,
Anglicised version) keeps their package updated, regularly refiles bugs
from them here if appropriate, and is otherwise one of most closely
cooperating distribution maintainers we have, so anyone running the Debian
packages is well received on the pan lists. You can rest easy he's keeping
in touch. =8^)
--
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
Pan-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/pan-users
--
Guilhem BONNEFILLE
-=- #UIN: 15146515 JID: [EMAIL PROTECTED] MSN: [EMAIL PROTECTED]
-=- mailto:[EMAIL PROTECTED]
-=- http://nathguil.free.fr/
Le script a débuté sur lun 16 oct 2006 21:46:19 CEST [EMAIL PROTECTED]:~$ pan --debug (article-cache.cc:175:ArticleCache) loaded 0 articles into cache from /home/peer/.pan2/article-cache (gnksa.cc:625:do_check_from) From [] (gnksa.cc:631:do_check_from) Name [] (gnksa.cc:632:do_check_from) Addr [] (gnksa.cc:638:do_check_from) GNKSA [102] (queue.cc:195:process_task) in process_task with a task of type GROUPS (nntp-pool.cc:208:request_nntp) trying to create a socket (queue.cc:195:process_task) in process_task with a task of type GROUPS (nntp-pool.cc:208:request_nntp) trying to create a socket (queue.cc:195:process_task) in process_task with a task of type GROUPS (nntp-pool.cc:208:request_nntp) trying to create a socket (queue.cc:195:process_task) in process_task with a task of type GROUPS (nntp-pool.cc:208:request_nntp) trying to create a socket (socket-impl-gio.cc:295:GIOChannelSocket) GIOChannelSocket ctor 0x83a1300 (socket-impl-gio.cc:295:GIOChannelSocket) GIOChannelSocket ctor 0x8467aa0 (nntp.cc:294:write_next_command) nntp 0x84702d0 writing to socket 0x83a1300 on server 4 this command: [] (socket-impl-gio.cc:517:set_watch_mode) socket 0x83a1300 calling set_watch_mode 1; _channel is 0x8472d30 (socket-impl-gio.cc:537:set_watch_mode) channel 0x8472d30 setting mode write (socket-impl-gio.cc:544:set_watch_mode) set_watch_mode 1: _tag_watch is now 86 (socket-impl-gio.cc:471:gio_func) gio_func: socket 0x83a1300, channel 0x8472d30, cond 4 (socket-impl-gio.cc:429:do_write) socket 0x83a1300 channel 0x8472d30 maybe wrote []; status was 1 (socket-impl-gio.cc:517:set_watch_mode) socket 0x83a1300 calling set_watch_mode 0; _channel is 0x8472d30 (socket-impl-gio.cc:530:set_watch_mode) channel 0x8472d30 setting mode read (socket-impl-gio.cc:544:set_watch_mode) set_watch_mode 0: _tag_watch is now 88 (socket-impl-gio.cc:451:do_write) channel 0x8472d30 done writing (socket-impl-gio.cc:495:gio_func) gio_func exit 0x83a1300 with retval 0 (socket-impl-gio.cc:295:GIOChannelSocket) GIOChannelSocket ctor 0x8461ad8 (nntp.cc:294:write_next_command) nntp 0x8475790 writing to socket 0x8467aa0 on server 5 this command: [] (socket-impl-gio.cc:517:set_watch_mode) socket 0x8467aa0 calling set_watch_mode 1; _channel is 0x8471138 (socket-impl-gio.cc:537:set_watch_mode) channel 0x8471138 setting mode write (socket-impl-gio.cc:544:set_watch_mode) set_watch_mode 1: _tag_watch is now 91 (socket-impl-gio.cc:471:gio_func) gio_func: socket 0x8467aa0, channel 0x8471138, cond 4 (socket-impl-gio.cc:429:do_write) socket 0x8467aa0 channel 0x8471138 maybe wrote []; status was 1 (socket-impl-gio.cc:517:set_watch_mode) socket 0x8467aa0 calling set_watch_mode 0; _channel is 0x8471138 (socket-impl-gio.cc:530:set_watch_mode) channel 0x8471138 setting mode read (socket-impl-gio.cc:544:set_watch_mode) set_watch_mode 0: _tag_watch is now 93 (socket-impl-gio.cc:451:do_write) channel 0x8471138 done writing (socket-impl-gio.cc:495:gio_func) gio_func exit 0x8467aa0 with retval 0 (nntp.cc:294:write_next_command) nntp 0x8475f78 writing to socket 0x8461ad8 on server 2 this command: [] (socket-impl-gio.cc:517:set_watch_mode) socket 0x8461ad8 calling set_watch_mode 1; _channel is 0x8475df8 (socket-impl-gio.cc:537:set_watch_mode) channel 0x8475df8 setting mode write (socket-impl-gio.cc:544:set_watch_mode) set_watch_mode 1: _tag_watch is now 101 (socket-impl-gio.cc:471:gio_func) gio_func: socket 0x8461ad8, channel 0x8475df8, cond 4 (socket-impl-gio.cc:429:do_write) socket 0x8461ad8 channel 0x8475df8 maybe wrote []; status was 1 (socket-impl-gio.cc:517:set_watch_mode) socket 0x8461ad8 calling set_watch_mode 0; _channel is 0x8475df8 (socket-impl-gio.cc:530:set_watch_mode) channel 0x8475df8 setting mode read (socket-impl-gio.cc:544:set_watch_mode) set_watch_mode 0: _tag_watch is now 103 (socket-impl-gio.cc:451:do_write) channel 0x8475df8 done writing (socket-impl-gio.cc:495:gio_func) gio_func exit 0x8461ad8 with retval 0 (socket-impl-gio.cc:295:GIOChannelSocket) GIOChannelSocket ctor 0x84762b8 (nntp.cc:294:write_next_command) nntp 0x8476320 writing to socket 0x84762b8 on server 1 this command: [] (socket-impl-gio.cc:517:set_watch_mode) socket 0x84762b8 calling set_watch_mode 1; _channel is 0x84761e0 (socket-impl-gio.cc:537:set_watch_mode) channel 0x84761e0 setting mode write (socket-impl-gio.cc:544:set_watch_mode) set_watch_mode 1: _tag_watch is now 106 (socket-impl-gio.cc:471:gio_func) gio_func: socket 0x84762b8, channel 0x84761e0, cond 4 (socket-impl-gio.cc:429:do_write) socket 0x84762b8 channel 0x84761e0 maybe wrote []; status was 1 (socket-impl-gio.cc:517:set_watch_mode) socket 0x84762b8 calling set_watch_mode 0; _channel is 0x84761e0 (socket-impl-gio.cc:530:set_watch_mode) channel 0x84761e0 setting mode read (socket-impl-gio.cc:544:set_watch_mode) set_watch_mode 0: _tag_watch is now 108 (socket-impl-gio.cc:451:do_write) channel 0x84761e0 done writing (socket-impl-gio.cc:495:gio_func) gio_func exit 0x84762b8 with retval 0 (queue.cc:195:process_task) in process_task with a task of type GROUPS (queue.cc:195:process_task) in process_task with a task of type GROUPS (queue.cc:195:process_task) in process_task with a task of type GROUPS (queue.cc:195:process_task) in process_task with a task of type GROUPS (data-impl.cc:101:save_state) data-impl dtor saving server, group, newsrc... Existing Quarks: 4: [1] (refcount 1) [2] (refcount 1) [4] (refcount 1) [5] (refcount 1)
[EMAIL PROTECTED]:~$ Script complété sur lun 16 oct 2006 21:46:55 CEST
_______________________________________________ Pan-users mailing list Pan-users@nongnu.org http://lists.nongnu.org/mailman/listinfo/pan-users
