[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[leafnode-list] 2.0b status
KNOWN BUGS
==========
1) "fetchnews -N" doesn't work. Reported by Juergen Salk <juergen.salk@xxxxxx>.
2) "fetchnews -S -M" led me once into an endless loop after the article
had been fetched for the first time. This possibly happens only if
the group is not marked "interesting". In that case, you can also
see some interesting buffer overflows in the Xref: header.
3) If a newsgroup occurs twice in an article header, the article is also
stored twice. Reported by Peter Holm <peter.holm@xxxxxx>.
4) The documentation is still incomplete. Most notably, the README misses
section 4.2 ("the filter file").
5) The timeout code for fetchnews still sucks (it has not been changed).
6) vsnprintf.c generates two warnings if both HAVE_SNPRINTF and HAVE_VSNPRINTF
are undefined.
7) Moderated groups are still not handled very well. Postings show up
on the server unmoderated; postings to local moderated groups don't
go to the moderators.
8) The groupinfo file is apparently not always updated properly. When
I subscribed to de.soc.weltanschauung.scientology, I could not read
any of the 300something new articles because they turned not up in
the groupinfo file.
9) If the complete grouplist is retrieved from a new server, the
new groups are not shown as new. Reported by Stefan Wiens
<s.wi@xxxxxxx>. Therefore some newsreaders (netscape, slrn) do
not find the group because they cache the list.
10) If the connection is not successful or interrupted, fetchnews
often hangs. If it receives a SIGINT then, it does not update
the groupinfo file.
11) Control messages are shown as well.
12) nntpd: postings in non-existent newsgroups are happily accepted
with 240. I have no idea what happens to them afterwards -
probably they get lost. (11Jan01)
13) The nntpd should include a Lines: header if there isn't one.
14) fetchnews may coredump on incorrect config files (i.e. if "kill"
was specified before "action" and "initialfetch" is set). Not
reproducible. Reported by Stefan Foerster
<foerstes@xxxxxxxxxxxxxxxxxxxxxxxxxxxx> on 15 Jan 2001.
15) fetchnews is unable to kill on header information which is not
present in the XOVER record (22 Jan 2001). Why?
16) delaybody mode does not work. Reported by Juergen Salk.
17) If a group is marked "interesting", it will go into the groupinfo file
during "fetchnews -f" even though it might not exist on the upstream
server any more. Reported by Udo Konstantin <Udo.Konstantin@xxxxxxxxxxx>
on 30 Jan 2001.
18) Groups that are not in the active file will never be deleted by
texpire. Is this a bug or a feature?
19) 120101: Wilhelm Meier <meier@xxxxxxxxxxxxxxxxxxx> & Stefan Wiens:
GROUP bla.bla
XOVER
POST into bla.bla
XOVER
the new posting doesn't show up. If a GROUP is inserted
before the XOVER command, it works.
20) When expiring the last article from a local group, the counters
are apparently still reset to 1.
POSSIBLY RESOLVED BUGS
======================
1) "make dist" fails at the moment when compressing the tarballs. I
have no idea why.
This is apparently a local problem and not reproducible by
others.
2) strdup.c doesn't compile and throws around a lot of odd error messages.
This may be a local problem. It does not occur when the
#include <string.h> line is commented out.
3) Articles are possibly lost when incorrectly linked (i.e. if
/var/spool/news/message.id/xxx/<bla@fasel> is already present
but /var/spool/news/name/of/group/nnnnn is not).
4) Memory leak in nntpd?
5) If storing local groups, the articles is not stored in
/var/spool/news/message.id/nnn if "nnn" does not exist. First reported
by Dennis W. Bulgrien <dbulgrien@xxxxxxxx> on 23 Jan 2001.
This problem should be fixed with 2.0b9.
--Cornelius.
--
/* Cornelius Krasel, U Wuerzburg, Dept. of Pharmacology, Versbacher Str. 9 */
/* D-97078 Wuerzburg, Germany email: krasel@xxxxxxxxxxxxxxxxxxxxxxxxxxxx */
/* "Science is the game we play with God to find out what His rules are." */
--
leafnode-list@xxxxxxxxxxxxxxxxxxxxxxxxxxxx -- mailing list for leafnode
To unsubscribe, send mail with "unsubscribe" in the subject to the list