Re: ERR_CANNOT_FETCH on DU 4.0 and Squid 1.1.8

From: Redfern Ian <RedfernI@dont-contact.us>
Date: Wed, 23 Apr 1997 21:30:00 +0100

I have single_parent_bypass on, inside_firewall set and my firewall
configured as parent no-query default.

I see this almost every weekday, for spells of 80 seconds, surrounded by
TCP connection failed/succeeded messages in cache.log. It doesn't
necessarily correspond to times of peak load on either machine - in fact
it's often late at night as well as lunchtimes.

I have amended peerCheckConnectStart() in my neighbors.c to say
eventAdd("peerCheckConnect", peerCheckConnect, p, 1);
so it will only be down for 1 second if this is a transitory thing, as
well as changing the error message in proto.c. The subsequent checks in
peerCheckConnectDone() are unaffected; I just want the first one to
happen immediately.

Ian Redfern (redferni@logica.com).
 ----------
From: Dave Hartley
To: Duncan Anker
Cc: squid-users@nlanr.net; tom@interact.net.au
Subject: Re: ERR_CANNOT_FETCH on DU 4.0 and Squid 1.1.8
Date: 23 April 1997 9:34am

I also changed the value but this didn't work. It may have ameliorated
the
problem though.

Dave

On 21-Apr-97 Duncan Anker wrote:
>I have changed the timeout from 2 to 10 seconds to see if that
>makes any difference.
>
>This is an intermittent problem, so I may not know if it has
>worked for a little while.

 ------

Dave Hartley Tel No: +44-1684-896000
DRA Malvern
St. Andrews Road Fax No: +44-1684-895700
Malvern
WR14 3PS email:
hartley@dra.hmg.gb
UK
Received on Wed Apr 23 1997 - 13:52:57 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:35:01 MST