[squid-users] The system returned: (111) Connection refused;

deepa ganu deepaganu at gmail.com
Fri May 27 15:03:11 UTC 2016


On May 27, 2016 5:43 PM, "Amos Jeffries" <squid3 at treenet.co.nz> wrote:
>
> On 27/05/2016 8:55 p.m., deepa ganu wrote:
> > Hi
> > I am using squid as a reverse.
> >
> > #http_port  80 accel defaultsite=202.53.13.19
> > https_port 443 accel  cert=/var/www/html/webrtc/imp/teleuniv.net.crt
> > key=/var/www/html/webrtc/imp/teleuniv.net.key
> > cafile=/var/www/html/webrtc/imp/intermediate.crt
defaultsite=202.53.13.19
> > no-vhost
> >
> >
> > #this ACL is url path specific which accepts only portal urls and deny
> > others.
> > acl portal urlpath_regex ^/portal6may
> > cache_peer 172.20.36.144 parent 80 0 no-query originserver
name=portalserver
> > cache_peer_access portalserver allow portal
> > cache_peer_access portalserver deny all
> > http_access allow portal
> >
> >
> > cache_peer 172.20.36.150 parent 443 0 no-query originserver ssl
> > sslflags=DONT_VERIFY_PEER login=PASS connection-auth=off name=teleuniv
> > acl our_sites dstdomain 202.53.13.19
> > http_access allow our_sites
> > cache_peer_access teleuniv allow our_sites
> > cache_peer_access teleuniv deny all
> >
> > SO when i try to access the url https://202.53.13.19/ I get the
following
> > error
> > "The following error was encountered while trying to retrieve the URL:
The
> > system returned: (111) Connection refused; The remote host or network
may
> > be down. Please try the request again."
> >
> > It only gives for 172.20.36.144 not for the urlpath acl.
>
> You have configured Squid to:
>
>  1) "no-vhost" - ignore the Host header the client sent indicating what
> domain name it was contacting.
It usually is vhost and the domain name is teleuniv.com but for testing
purposes we use public IP 202.53.13.19
>
>  2) defaultsite=202.53.13.19 - use "202.53.13.19" as the domain *name*
> for all requests received through that https_port.
>
> Why would you expect to see anything other than https://202.53.13.19 in
> the URL when you have configured those?
Didnt understand the above mentioned point
>
> > But this happens
> > only sometimes. When I physically go to that server (172.20.36.150) and
> > click on the wired connection (one of the LAN options using linux). It
> > works again. I am very confused
>
> Your use of "physically" seems to be incorrect. You walked up to the
> machine hardware and did what ?
> I logged in the machine and click on "wired connection"
> "Clicking" seems to be that you logged in (not physically) and changed
> something which affected how Squid was able to connect to it.
>
>
> All traffic with the domain name "202.53.13.19" and not the path
> "/portal6may" gets sent to the second cache_peer (172.20.36.150).
>
> So what do you expect to happen when the server 172.20.36.150 receives a
> request with Host: header domain name set to "202.53.13.19"  ?
>I didn't understand the host: header part any references to understand
>
> It seems like the servers connectivity is a bit flakey and getting
> disconnected occasionally. But the wrong Squid configuration could be
> hiding some other issue.
> I read somewhere it happens when cache is full. Can u tell me what should
be cache size if my HDD 500 gb and 32 MB RAM
> Amos
>
> _______________________________________________
> squid-users mailing list
> squid-users at lists.squid-cache.org
> http://lists.squid-cache.org/listinfo/squid-user
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.squid-cache.org/pipermail/squid-users/attachments/20160527/31e62ef1/attachment.html>


More information about the squid-users mailing list