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

Re: [leafnode-list] groups listed twice in groupinfo, and other



Jan Knutar <jknutar@xxxxxx> writes:

> Ok, xxd produced the most usable output imo:
>
> Example group alt.sci.seti
>
> Commandline: grep alt.sci.seti groupinfo|xxd
>
> 0000000: 616c 742e 7363 692e 7365 7469 2037 3637  alt.sci.seti 767
> 0000010: 3430 2032 2031 3033 3630 3730 3433 3620  40 2 1036070436 
> 0000020: 5365 6172 6368 2066 6f72 2045 7874 7261  Search for Extra
> 0000030: 2054 6572 7265 7374 7269 616c 2049 6e74   Terrestrial Int
> 0000040: 656c 6c69 6765 6e63 650a 616c 742e 7363  elligence.alt.sc
> 0000050: 692e 7365 7469 2037 3635 3837 2032 2031  i.seti 76587 2 1
> 0000060: 3033 3630 3730 3433 3620 5365 6172 6368  036070436 Search
> 0000070: 2066 6f72 2045 7874 7261 2054 6572 7265   for Extra Terre
> 0000080: 7374 7269 616c 2049 6e74 656c 6c69 6765  strial Intellige
> 0000090: 6e63 650a                                nce.

So we have real double entries. The group arrived recently, the fourth
column denotes when it has arrived in your setup, decoding it to GMT:

$ perl -MPOSIX -e 'print scalar gmtime(1036070436), "\n";'
Thu Oct 31 13:20:36 2002

Did you run fetchnews -f at that time? If yes, then something is really
hosed. If fetchnews ran without -f, then I need to look for a bug.

> This from Sunday's backup of the newsspool, so ok, the duplication of 
> groupinfo lines is very recent, either when I upgraded to leafnode 
> .29, or the patch v1/v2 (nah not possible?)...

No, that patches don't to that.

> *reaches for cron and stops teh fetchnews job*
>
> I seem to recall that I did touch /var/spool/news/active.read and chown 
> news.news /var/spool/news/active.read to stop fetchenws from 
> re-acquiring the kust if newsgroups over and over again (this was 
> eating lots of bandwidth)... Could this have triggered or caused the 
> duplicate groupinfo lines I'm seeing now? (active.read file was 
> nonexistant before i touch'd it, /var/spool/news has ownership 
> news.news and drwxrwsr-x)

That would have quelched fetchnews' -f option, but should not (at least
not by itself) cause duplicate news groups.

Does running fetchnews -f fix the problem? Please tell me whether that
fetchnews -f run reached all servers or if any one could not be
connected to or broke the connection in "mid-flight".

-- 
Matthias Andree

http://sourceforge.net/projects/leafnode/   - leafnode-1 tarballs
http://mandree.home.pages.de/leafnode/beta/ - leafnode-2 tarballs

-- 
leafnode-list@xxxxxxxxxxxxxxxxxxxxxxxxxxxx -- mailing list for leafnode
To unsubscribe, send mail with "unsubscribe" in the subject to the list