Re: [squid-users] parent request order

Hey TLS.

There is a solution for that of-course.
you should use some External_Acl helper that will check the parent
proxy(each and one of them) for a test like "www.example.com"
which the external_acl should receive some specific data and that it
will not be served from cache...
take a look at my code in:
https://github.com/elico/squid-helpers/tree/master/squid_helpers/proxy_hb_check

This is an example that replaces the old cache_peer "monitorX"
http://www.squid-cache.org/Doc/config/cache_peer/
The "monitorX" that was yet to be ported into 3.3 and 3.head should do
exactly what you need.

So for now use external_acl helper that will make sure which of the
proxies is on and will mark it as on.
the external helper should check for the parent state and if both are
not up you can trigger some alarm or something else.

Do you think this kind of solution will fit your scenario?

Eliezer

On 06/24/2013 12:01 PM, 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 I want to do is:
>>> >a) switch to alternative parents automatically
>>> >b) in case of P_1-failure: distribute N_1-traffic over S_2 and S_3
>>> >c) in normal operation send all requests to P_1 and no requests to
>>> S_2 or
>>> >S_3
>>>
>>> wpad does a good job.
>>> I am not sure how it balances but the failover works fine.
>> ...
>> So, what you can do is divide the client IP address range in
>> 3 segments and via the PAC file make a decision which
>> proxy (S_i) will be the primary proxy; the browser wil only
>> use the secondary proxy when the first one is not available.
>
> I checked this out, but it did not work, because when the PAC-File returns:
>
> "PROXY p1:8080; PROXY p2:8080";
>
> and the parent-proxy of p1 is not available, p1 returns a error-message
> ("parent not available") and the browser will not failover to p2,
> because the request yield HTML and everything seems to be fine for the
> browser.
>
> I am not sure, if turning off the error massages is an option?
>
> Kind regards
> Thomas
Received on Mon Jun 24 2013 - 11:48:07 MDT