Re: [squid-users] One squid instance, two WAN links, how to failover from primary to secondary link?

From: Amos Jeffries <squid3_at_treenet.co.nz>
Date: Tue, 27 Aug 2013 17:55:45 +1200

On 27/08/2013 5:09 a.m., Thomas Harold wrote:
> On 8/26/2013 6:41 AM, Nishant Sharma wrote:
>> Hi Thomas,
>>
>> Thomas Harold <thomas-lists_at_nybeta.com> wrote:
>>> In an instance where you have a single instance of squid running on
>>> a two WAN links as WAN
>>> #2
>>> is very slow compared to WAN #1.
>>>
>>> Is this simply handled by changing the default gateway of the server
>>> using the "ip route" commands when we detect that WAN#1 is down?
>>
>> Yes, it should work that way. Simple and easy.
>>
>
> I'm guessing that "balance_on_multiple_ip" should be set to "off" as
> well, to keep squid from balancing across both WAN links?

balance_on_multiple_ip has nothign to do with Squid and WAN links. It is
"balancing" requests over the different remote IPs when a website lists
more than 1 IP in its DNS results.
Otherwise known as "client-based load balancing".

>
> http://www.squid-cache.org/Doc/config/balance_on_multiple_ip/
>
> It's not clear what the manual means when it says "Modern IP resolvers
> in Squid sort lookup results by preferred access. By default Squid
> will use these IP in order and only rotates to the next listed when
> the most preffered fails."
>
> What is "preferred access"? Is that defined somewhere? Or does that
> mean the preferred routes defined by the TCP/IP network stack?

http://www.squid-cache.org/Doc/config/dns_v4_first/

Amos
Received on Tue Aug 27 2013 - 05:55:52 MDT

This archive was generated by hypermail 2.2.0 : Tue Aug 27 2013 - 12:00:26 MDT