On 06/24/2013 12:44 PM, T Ls wrote:
Am 24.06.2013 13:08, schrieb Marcus Kool:
On 06/24/2013 06:01 AM, T Ls wrote:
Am 19.6.2013 16:13, schrieb Marcus Kool:
On Wed, Jun 19, 2013 at 09:27:54AM -0300, Marcus Kool wrote:
On 06/19/2013 09:02 AM, T Ls wrote:
>...
What means "not available" ??
If proxy p1 is "not available", usually the browser cannot connect to it
or a request times out.
With a no-connect error, a browser does failover to the second proxy...
Just checking: p1 and p2 are parent proxies and s1,s2,s3 the proxies
that the serve the browser requests?
If yes, shouldn't the PAC code return "PROXY s1:8080; PROXY s2:8080;
PROXY s3:8080" ?
Yes of course, I'm sorry, I mixed up the names. My PAC returns
"PROXY s1:8080; PROXY s2:8080; PROXY s3:8080"
for N_1, and S_1 hands the request over to P_1, if P_1 is dead (that is currently the case) S_1 returns an error page (HTML) to the client in N_1, saying that the parent proxy is not responding.
Sorry for the confusion.
Thomas
Yeah this is not what you want.
For proxy failover by PAC to work, it is desired that Squid S1 does not send
an HTTP message with an error but disconnects to make the browser go to S2,
or better does not accept connections when its only parent is not available...
Can't you make a setup where S1 uses P2 if P1 fails?
In an old thread I read Squid has a configuration option FIRST_UP_PARENT
so it can be configured to use P1 and only P2 if P1 is not available.
Marcus
