On Thu, 13 Nov 1997, Oskar Pearson wrote:
> > I've just found the RETRY_CONNECT option in comm.c and squid is now
> > running with that enabled. It seems to fix the problem, but why are
> > connects to sites failing?
> Problem is, however, that squid doesn't do this properly (there are
> patches, but I don't know how stable the RETRY_CONNECT is). If one
> of the IP's is down, the site is down.
This not the case though. e.g. I've just had the following in my
cache.log:
97/11/13 12:38:08| connect: 166.90.104.136:80: (32) Broken pipe.
97/11/13 12:38:08| Retrying connection to www.xoom.com
However, from the cache machine I can connect just fine:
# telnet 166.90.104.136 80
Trying 166.90.104.136...
Connected to 166.90.104.136.
Escape character is '^]'.
The connect call is setting errno to 32, which isn't possible according to
the manual pages? I'm thinking this might be Solaris problem.
> This is fixed proper with squid-1.2 and various patches are available
> for 1.1
>
> Oskar
Mark Powell - System Administrator (UNIX) - Clifford Whitworth Building
A.I.S., University of Salford, Salford, Manchester, UK.
Tel: +44 161 295 5936 Fax: +44 161 295 5888
Email: M.S.Powell@ais.salford.ac.uk finger mark@ucsalf.ac.uk (for PGP key)
Received on Thu Nov 13 1997 - 04:43:07 MST
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:37:31 MST