Re: Dynamic content policy update

Ah, missed the refresh_pattern -- sorry!

On 29/01/2008, at 2:01 PM, Amos Jeffries wrote:

> Mark Nottingham wrote:
>> Well, RFC2616 section 13.9;
>>> We note one exception to this rule: since some applications have
>>> traditionally used GETs and HEADs with query URLs (those
>>> containing a "?" in the rel_path part) to perform operations with
>>> significant side effects, caches MUST NOT treat responses to such
>>> URIs as fresh unless the server provides an explicit expiration
>>> time. This specifically means that responses from HTTP/1.0 servers
>>> for such URIs SHOULD NOT be taken from a cache.
>> ... so technically you'd be non-conformant regarding the MUST NOT
>> (unless there's code to back this up).
>
> Read that again :-)
>
> At present default policy these URI are NEVER cached.
>
> The altered policy would enable squid to work with the noted
> exception: "unless the server provides an explicit expiration time".
> The zeros in the pattern prevent freshness, yes?
>
> It only breaks the SHOULD NOT about data from 1.0 servers.
>
> The MUST would be kept _exactly_ to the _full_ wording without
> administrator intervention.
>
> Amos
>
>
>> On 29/01/2008, at 12:53 PM, Amos Jeffries wrote:
>>> Any objections to Squid2-HEAD and Squid 3-HEAD getting the caching
>>> policy adjustment for dynamic pages Adrian proposed before xmas?
>>>
>>> Namely Dropping the
>>> acl QUERY cgi-bin \?
>>> cache deny QUERY
>>>
>>> Replacing it with:
>>> refresh_pattern (cgi-bin|\?) 0 0% 0
>>>
>>>
>>> No terrible side-effects here so far after a month of use. Just
>>> the bugzilla Expires bug coming to view and being fixed.
>>>
>>> Amos
>>> --
>>> Please use Squid 2.6STABLE17+ or 3.0STABLE1+
>>> There are serious security advisories out on all earlier releases.
>> --
>> Mark Nottingham mnot@yahoo-inc.com
>
>
> --
> Please use Squid 2.6STABLE17+ or 3.0STABLE1+
> There are serious security advisories out on all earlier releases.

--
Mark Nottingham       mnot@yahoo-inc.com