Steve A wrote:
> Sorry I wasn't more helpful.
Don't be! As I said before, talking through seems to help pin down the
misbehaviour.
> I assume you've played with the leafnode config option for slow upstream
> servers ? One should be able to increase the time out variable.
At the moment I have this in /etc/news/leafnode/config, posted by someone in
my ISP's Linux newsgroup (not disseminated):
server = news-text.blueyonder.co.uk
expire = 40
timeout = 25
initialfetch = 200
server = news.blueyonder.co.uk
expire = 40
timeout = 25
initialfetch = 200
He says it works for him but some weird crap is happening on my machine.
The processes started by the hourly cron job never finish. For example, the
10:08 cron job sent this mail "Cannot obtain lock file, aborting." because
the 09:08 job is still running ("ps aux" shows that these started at 09:08)
/home/adam $ ps ax|grep news
685 ? Ss 0:00 /bin/sh -c if
[ -x /etc/news/leafnode/do-fetch-news ];
then /etc/news/leafnode/do-fetch-news; fi
687 ? S 0:00 /bin/sh /etc/news/leafnode/do-fetch-news
689 ? S 0:01 /usr/sbin/fetchnews
although the processes don't appear to be doing anything:
"find /var/spool/news -mmin -60" returns only the /var/spool/news/leaf.node
directory.
"grep news /var/log/syslog" also shows nothing happening between 09:09 and
the 10:08 run failing to get the lock.
...
Jun 1 09:09:09 garcia fetchnews[689]: sci.archaeology: killed 37036
(<[EMAIL PROTECTED]>), already fetched
before
Jun 1 09:09:09 garcia fetchnews[689]: sci.archaeology: killed 37090
(<[EMAIL PROTECTED]>), already fetched before
Jun 1 10:08:01 garcia /USR/SBIN/CRON[1818]: (news) CMD (if
[ -x /etc/news/leafnode/do-fetch-news ];
then /etc/news/leafnode/do-fetch-news; fi)
Jun 1 10:08:01 garcia fetchnews[1822]: leafnode 1.11.2.rel: verbosity level
is 0, debugmode is 0
Jun 1 10:08:06 garcia fetchnews[1822]: Cannot obtain lock file, aborting.
-- Adam
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
