Re: request

Sandeep Kuttal wrote:
> Hi Amos,
>
> Thanks for showing interest in my research. I am looking for
> improving on a site for that I observed the data using httpfox i.e
> extension of firefox to study data going and coming. After that I got

Ah, so what you are looking at is not in fact caching of POST data but
how the site needs to be optimized for reduced traffic?

> to know that most of the important data intended for my research was
> in POST messages both reply and request.

So for optimization the question there becomes: is POST the right
request type to be doing each of the actions?

GET, POST, PUT, and other request types each has a fixed set of defined
meanings. That defined meaning is also the reason why some requests can
be cached and some cannot.

In my experience as a website developer, I've seen many dynamic sites
using POST where they should be using GET.

> So I intended on using a
> cache server to capture the data and save it. The best cache server
> out there is squid. So I went through the squid and found out that it
> stores only GET message. I was looking for archives of squid to get a
> view how to save the POST messages but till now got no success.
> According to archives there are two solutions:
>
> 1) write a small ICAP servlet implementing this cache

This is the easy one for monitoring the data. It's relatively hard to
identify and pull things out of the squid cache.

>
> 2) * Modify how request entities are forwarded, temporarily
> buffering the entity so you can process the request entity before the
> request is forwarded. * Extend the public storeKey hashed data with
> information as required to give each POST request/reply which should
> be cached uniquely a unique storekey. * Remove the flag which makes
> POST requests uncacheable (look for METHOD_POST).
>
> Some body suggested me to look at ICAP/e-CAP and somebody suggested
> to look at the second approach. So I am stuck how to start with. I
> don't have any clue from where should I start. Kindly guide me
> through this. Since this is the most important part of my research to
> capture these POST messages.
>
> Thanks Sandeep
>
>
>
>
> On Apr 16, 2010, at 10:36 PM, Amos Jeffries wrote:
>
>> Sandeep Kuttal wrote:
>>> Hi, I am a PhD student working on research for that I need to
>>> cache the POST messages and I know no proxy server allows to
>>> cache that. I want to work directly on it I can modify the squid
>>> code some how and let it cache POST messages just for my research
>>> since according to RFC the cache servers won't allow to cache
>>> POST messages. Kindly subscribe me to your development list.
>>> Thanks, Sandeep
>> Greetings Sandeep,
>>
>> Firstly welcome. It is always good to see someone new interested in
>> Squid.
>>
>>
>> I have some questions about your intended research;
>>
>> What results you do see as being gained from this?
>>
>> Have you read up on the reasons why the RFCs says POST cannot be
>> cached in shared proxies such as Squid? (the emphasis being on
>> *shared*)
>>
>> Which part(s) of the POST->reply sequence do you see as being of
>> any use caching?
>>

Amos

-- 
Please be using
   Current Stable Squid 2.7.STABLE9 or 3.1.1