[leafnode-list] Failed posting - is this right?

John Carlyle- Clarke john.cc at europlacer.co.uk
Tue Jun 22 10:34:03 CEST 2004


Running 2.0.0.alpha20040513a...

I fetch from several servers, all but one of which carry only a small number of groups.  I noticed that when I sent a post to the server during a fetch, it looks like leafnode tried to send the article to the server it was currently fetching, even though that server did not carry the right group...

Here is the section of the log file:

Jun 21 10:00:59 pc69 fetchnews[26690]: news.jrsoftware.org: connecting to port nntp
Jun 21 10:00:59 pc69 fetchnews[26690]:   trying:    address 66.98.148.123 port 119...
Jun 21 10:00:59 pc69 fetchnews[26690]:   connected: address 66.98.148.123 port 119.
Jun 21 10:00:59 pc69 fetchnews[26690]: news.jrsoftware.org: connected (200), banner: "200 news.jrsof
tware.org InterNetNews NNRP server INN 2.4.1 ready (posting ok)."
Jun 21 10:01:00 pc69 fetchnews[26690]: news.jrsoftware.org: checking for new newsgroups
Jun 21 10:01:00 pc69 fetchnews[26690]: news.jrsoftware.org: found 0 new newsgroups
Jun 21 10:01:00 pc69 fetchnews[26690]: found 0 articles in out.going.
Jun 21 10:01:00 pc69 fetchnews[26690]: news.jrsoftware.org: 0 articles posted
Jun 21 10:01:04 pc69 fetchnews[26690]: jrsoftware.innosetup: no new articles
Jun 21 10:01:05 pc69 fetchnews[26690]: jrsoftware.innosetup.thirdparty: no new articles
Jun 21 10:01:05 pc69 leafnode[26677]: UPSTREAM POST <Xns950F702713718johncceuroplacercouk at 192.168.1.
69> borland.public.delphi.com.activex.writing
Jun 21 10:01:06 pc69 fetchnews[26690]: Failed to post /var/spool/news/out.going/26677-1087812065-1, 
moving to /var/spool/news/failed.postings/
Jun 21 10:01:07 pc69 leafnode[26697]: Cannot obtain lock file to store article /var/spool/news/temp.
files/26677-1087812065-1. It will be posted later.
Jun 21 10:01:08 pc69 fetchnews[26690]: wrote groupinfo with 38231 lines.
Jun 21 10:01:08 pc69 fetchnews[26690]: fetchnews: 95 articles and 0 headers fetched, 2 killed, 0 pos
ted, in 68 seconds

This server is the last in the list, and this is the first time the outgoing post has been mentioned.

The other possible explanation I can see is that I have the following cron job, which I got from this list at some point:

# Run a queue check and post outgoing every 5 minutes
*/5 * * * * /var/spool/news/qcheck

(Fetchnews is set to run every hour on the hour exactly.)

Here is /var/spool/news/qcheck:

#!/bin/sh
if [ -n "`find /var/spool/news/out.going/ -type f -print`" ]
then
        fetchnews -Pv
fi

Could this script have tried to run while fetchnews was already running?  I thought the lockfile would prevent this.

Hope someone can cast some light on this for me :)

Regards,

	John Carlyle-Clarke



More information about the leafnode-list mailing list