it's a customized tunnel can only transport TCP data.
on that diagram, SQUID BCD is actually act as a service cluster, for
many client like A. the data between A and BCD must be encrypt, and
the latency is expected to be pretty high on the link. flooding the
limited bandwidth tunnel with HTCPs probably not a good idea either.
However, per you, SQUID will track HTTP connect time and prefer faster
one by default? can it track "response time" instead? which means
time of a header exchange is done (which pretty much equals to the
link speed).
But I don't understand how SQUID reset the peer preference though, in
this case, if one of the peer is slow, squid wills till periodically
query it just to make sure it is still slow, but that will be very bad
experience for the end-user though.
Cheers.
On Tue, Mar 29, 2011 at 2:56 AM, Amos Jeffries <squid3_at_treenet.co.nz> wrote:
> On 29/03/11 21:39, Yucong Sun (叶雨飞) wrote:
>>
>> Hi there,
>>
>> So I have a squid a configured with several remote parent, with no ICP
>> port available (as the http port is tunneled in between)
>>
>> I think ICP serves both QUERY and HEALTH CHECK function in squid, but
>> I can't tunnel udp packets
>>
>> SQUID A -> SQUID B
>> -> SQUID C
>> -> SQUID D
>>
>> Due to the fact that the link in between is all different and possibly
>> in-stable from time to time, I would want A to try to fetch from
>> multiple peers at once and use which ever returns first.
>>
>> Could this be done?
>>
>> A alternative is to have squid detect dead/slow peer, I know this is
>> porbably done in ICP, but I can't tunnel udp packets.
>
> What type of tunnel is this that cannot transfer a particular protocol?
>
> Squid takes measures of HTTP connect times, ICP response times, HTCP
> response times and if the "pinger" is installed ICMP active destination
> probing. All of these affect the RTT measure on a cache_peer.
>
> Squid *will not* fetch from multiple peers at once. HTTP does not permit it
> by design. ICP and HTCP probes are used in parallel to determine if a peer
> is worth contacting. In their absence peers are tried in an sequential order
> assigned by the selection algorithm.
>
> Amos
> --
> Please be using
> Current Stable Squid 2.7.STABLE9 or 3.1.11
> Beta testers wanted for 3.2.0.5
>
Received on Tue Mar 29 2011 - 11:59:01 MDT
This archive was generated by hypermail 2.2.0 : Tue Mar 29 2011 - 12:00:02 MDT