Hi Russ,
By default, when this happens, innd will defer the article from the second peer (using a 436 response) rather than accept it from both peers. I
forget how that's counted in the statistics (if it is at all).
It is shown in the innd status file:
news.gegeweb.org
ip address: 2001:41d0:2:abb2::119
seconds: 28410 duplicates: 0 max allowed cxns: 8
offered: 318 uw newsgroups: 0 active cxns: 1
accepted: 28 uw distributions: 0 sleeping cxns: 0
refused: 290 unapproved: 0 want streaming: Yes
rejected: 0 filtered: 0 is streaming: Yes
size: 55.6kb bad sites: 0 duplicate size: 0.0kb
reject size: 0.0kb
Protocol:
Ihave: 0 SendIt[335]: 0 Got[435]: 0
Deferred[436]: 0
Check: 380 SendIt[238]: 28 Got[438]: 290
Deferred[431]: 62
Takethis: 28 Ok[239]: 28 Error[439]: 0
Maybe the term "Error" for the 439 response code to TAKETHIS is
misleading. I've seen it for:
Dec 3 05:08:15.137 - feeder.eternal-september.org <
[email protected]> 439
User-issued spam cancel
Probably "Rej[439]" would be better ("Rejected" is a bit too long,
unless adding more spaces).
--
Julien ÉLIE
« Bibere humanum est, ergo bibamus. »
--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)