2.3: "paused" nntp channels

Heath Kehoe heath.kehoe at intermec.com
Thu Apr 13 07:49:02 UTC 2000


>When I do:   ctlinnd name ''
>
>I get: 
>
>remconn:4:remconn::
>control:50:control::
>localconn:52:localconn::
>nntp-out4.uu.net:55:nntp:371:paused
>nntp-out4.uu.net:56:nntp:0:paused
>innfeed!:57:proc:19087:
>nntp-out4.uu.net:59:nntp:0:paused
>nntp-out4.uu.net:60:nntp:0:paused
>news-out.cwix.com:72:nntp:0:paused
>news-out.cwix.com:73:nntp:0:paused
>news-out.cwix.com:74:nntp:0:paused
>news-out.cwix.com:75:nntp:0:paused
>news-out.cwix.com:76:nntp:0:paused
>inflow:78:file::
>inpaths:80:file::
>controlchan!:82:proc:19089:
>pathmap:83:proc:21956:
>
>Why do all of the nntp sites say 'paused'?  Looking at the code, it
>prints 'paused' if (cp->MaxCnx > 0 && cp->ActiveCnx == 0) .  Which
>implies that cp->ActiveCnx is 0 for all of the sites.  It seems to me
>that this is incorrect.
>
>My incoming.conf has max-connections = 8 for all peers, and no setting
>of hold-time.
>


Ok, I'm closer to a solution.  I modified cc.c to print some more
info for nntp channels, and found that the cp->HoldTime settings
were garbage; they looked like uninitialized values.  So I put in
my incoming.conf:  "hold-time: 0"  to explicitly set HoldTime, and
that fixed the HoldTime settings; and also the "paused" problem.

So, the root cause is: HoldTime is not getting set to 0 by default.
Someone more familiar with RCreadfile() than I am needs to fix that.

-heath




More information about the inn-workers mailing list