[squid-users] rep_header not working

Amos Jeffries squid3 at treenet.co.nz
Tue Jul 19 13:30:51 UTC 2016


On 19/07/2016 10:19 p.m., Omid Kosari wrote:
> 
> If i understand correctly you mean the rule should work correctly with
> custom headers but the problem is squid is not at right place to see that
> header .
> 

Correct.

> May i ask you please help me to solve problem from other thread
> http://squid-web-proxy-cache.1019090.n4.nabble.com/Windows-Updates-a-Caching-Stub-zone-A-windows-updates-store-td4678454.html
> 

Sure, if I can assist with that I will do so in a reply to that thread.

> I think now you know what is my problem . The prefered way is by rep_header
> and clientside_tos if possible . 

It is not possible with rep_header ACL.

> Right now with help of Eliezer i have injected custom header in static
> header files . Eliezers code (peer port 8080) successfully sends that header
> to clients and squid(i don't know how to be sure ,the important if).
> 
> Is it possible to use rep_header and clientside_tos with each other ? (Alexa
> says no in other thread but he is not deeply aware of my needs ) 

Alex doesn't need to understand what you want to know that combining
those two particular things is impossible. For the reason he stated:
that clientsie_tos is evaluated _only_ at one point well before any
rep_header data is available.

Understanding your need/want/policy requirement is to figure out if
there is some alternative way we know that you might have missed.


> If yes how to squid be aware of rep_header from peer ?

There is not much Squid does once it has the reply. Just
http_reply_access, store_miss, and reply_header_access/replace/add at
the point of delivery to the client. I may have missed some uncommon
one(s), but IIRC everything else is either about the HTTP request or
handling the cleanup when finished (eg logging).

> 
> Knocking my head to wall :(
> 

:-(

Amos



More information about the squid-users mailing list