Announcements and discussions for Gnus, the GNU Emacs Usenet newsreader
 help / color / mirror / Atom feed
* Re: Stale connection?
       [not found] <m28ywty0ro.fsf@bluesteel.grierwhite.com>
@ 2003-02-06 15:03 ` Kai Großjohann
       [not found]   ` <m2ptq5ja8z.fsf@bluesteel.grierwhite.com>
  2003-02-07  2:13 ` Chris B.
  1 sibling, 1 reply; 3+ messages in thread
From: Kai Großjohann @ 2003-02-06 15:03 UTC (permalink / raw)


chris@grierwhite.com (Christopher J. White) writes:

> Any way to have gnus automatically close the connection after a
> timeout and then re-establish it as necessary?

This depends on how you connect to the server.
-- 
A turnip curses Elvis


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Stale connection?
       [not found] <m28ywty0ro.fsf@bluesteel.grierwhite.com>
  2003-02-06 15:03 ` Stale connection? Kai Großjohann
@ 2003-02-07  2:13 ` Chris B.
  1 sibling, 0 replies; 3+ messages in thread
From: Chris B. @ 2003-02-07  2:13 UTC (permalink / raw)


chris@grierwhite.com (Christopher J. White) moved upon the face of the 'Net and spake thusly:

> If I leave gnus up and running for a while, then check news
> (g from group buffer: gnus-group-get-new-news), it frequently
> hangs as if waiting for a timeout.  I've found, however, that
> if I go unplugged (J j), then plugged again (J j), then 
> check news, it is nice and fast.  Is this normal behaviour?
> If I leave gnus plugged for a significant time, is gnus
> keeping the connection open on my end but the server
> is terminating it, thus the long timeout?  Any way to have
> gnus automatically close the connection after a timeout
> and then re-establish it as necessary?

I get that happen to me, because I connect to the NNTP server through
a NAT firewall.

The firewall box flushes its NAT tables after, I think, 10 minutes of
inactivity, which can easily be the time it takes to compose a
detailed followup.

I habitually do "C-x 5 b *Group* ^ C-n C O C-x 5 0" (pull up *Server*
window, close and re-open NNTP connection) to recover from this state.

It would be real nice if there were an easier way to detect/recover
from stale NNTP connections.  (Yeah, the proper solution is to
eradicate NAT, which is evil and loathesome).

--cjb


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: Stale connection?
       [not found]   ` <m2ptq5ja8z.fsf@bluesteel.grierwhite.com>
@ 2003-02-07 18:27     ` Kai Großjohann
  0 siblings, 0 replies; 3+ messages in thread
From: Kai Großjohann @ 2003-02-07 18:27 UTC (permalink / raw)


chris@grierwhite.com (Christopher J. White) writes:

> cj> Any way to have gnus automatically close the connection after a
> cj> timeout and then re-establish it as necessary?
>
> kai> This depends on how you connect to the server.
>
> I use nntp:

Hm.  I'm afraid that there might not be an easy solution.  Though, in
a way, the solution should be easy, if you are a Lisp hacker: add a
variable containing a timestamp for sending the most recent command.
Then, inside nntp-send-command, look for that timestamp.  And if more
than N seconds have passed since the last command, tear down the
connection and open a new one.
-- 
A turnip curses Elvis


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2003-02-07 18:27 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
     [not found] <m28ywty0ro.fsf@bluesteel.grierwhite.com>
2003-02-06 15:03 ` Stale connection? Kai Großjohann
     [not found]   ` <m2ptq5ja8z.fsf@bluesteel.grierwhite.com>
2003-02-07 18:27     ` Kai Großjohann
2003-02-07  2:13 ` Chris B.

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).