[leafnode-list] Re: strange problem

Matthias Andree matthias.andree at gmx.de
Tue Apr 24 09:10:32 CEST 2007


Nick C schrieb:
> On 18.03.2007 13:09, Nick C wrote:
> 
>> I have 2 nntp servers in config file. news.individual.net and
>> news.mozilla.org. The mozilla nntp server is fetched last in the process.
> 
>> Have a look at the timestamp (30 min distance). At 11:30 it starting fetch
>> process, but as you can see it does not finish it. It does not write down
>> how many articles it has fetched in this process.
> 
>> There is no line like this in log:
>> Mar 18 10:00:18 ubuntu fetchnews[5051]: fetchnews: 20 articles and 0 headers
>> fetched, 0 killed, 0 posted, in 15 seconds
> 
> It has started again, now with the news.individual.net nntp server. It
> starts fetchnews process, but does not finish it. It just stops on some
> random newsgroup. This only happens if the fetchnews is called by crontab
> (every 30 min). If I run fetchnews by hand from terminal, everything is fine.

Still very strange.

Can you add -vvvv to your crontab line, add debugmode=1 to your
configuration file and see what the last few lines of your log are?

Can you then also check (find /var/spool/news -name '*core*') if there
are any core files

It will grow your log tremendously, so once you've got some logging that
lacks the final line, you'll probably like to remove those -vvvv again
and set debugmode=0.

Mail me the resulting log and the output of "leafnode-version" off-list
(you can gzip the log if you like and you will have to if it's in excess
of a few MB).

Thanks
Matthias



More information about the leafnode-list mailing list