Re: [AMaViS-user] clarify: SQL lookups for local domains

[Jo R. <jr...@ne...>]

Thanks for the very clear answer.  I was pretty sure I'm on the right  
track, but I want to be clear about this.

How do you feel about documentation?  Willing to accept some  
documentation changes to make this easier to grok?

On Jan 30, 2008, at 8:32 AM, Mark Martinec wrote:
>> I have asked repeatedly on the list about how to query SQL for
>> local_domains_maps.  To first some answers which didn't seem to  
>> relate,
>> then resounding silence.
>>
>> After taking some time to read the code today, I believe I have  
>> figured
>> out the answer.  Will someone please tell me if I'm right or wrong?
>>
>> 1. There is no way to query SQL for local_domains
>
> Yes, kind of.  The locality flag is just one of the fields in users  
> table.
>
>> *HOWEVER*
>>
>> 2. When a recipient lookup is done in SQL, if the local column  
>> evaluates
>> (is not undefined) then the recipient is considered local.
>
> More exact: if the field evaluates to true, recipient is considered  
> local.
> If it evaluates to false, recipient is not considered local.
> If the field value is NULL (=undef), the lookup falls back to
> static settings.
>
> If there is no field 'local', then a default applies:
> a matching 'users' record suffices to consider recipient local,
> UNLESS the matching record has '@.' in its users.email field.
>
>> *AND*
>>
>> 3. A wildcard entry of '@{domain}' will evaluate the entire domain  
>> to be
>> local (unless you set $sql_lookups_no_at_means_domain in which  
>> case drop
>> the @ sign)
>
> Yes.
>
>> Therefore, the right answer for hosting environments is to make sure
>> that every domain you are receiving mail for exists in the users  
>> database.
>
> Exactly.
>
>> Questions:
>>
>> Is this analysis correct?
>
> Yes.
>
>> Would it be too much trouble to allow a SQL query to populate
>> local_domains_maps, just like the hash lookup.
>
> That would require another kind of a SQL lookup mechanism,
> not in use for any other setting. Currently I think the
> unified SQL lookups mechanism is general enough and suffices
> for all present needs.
>
>   Mark
>
> ---------------------------------------------------------------------- 
> ---
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> AMaViS-user mailing list
> AMa...@li...
> https://lists.sourceforge.net/lists/listinfo/amavis-user
> AMaViS-FAQ:http://www.amavis.org/amavis-faq.php3
> AMaViS-HowTos:http://www.amavis.org/howto/

-- 
Jo Rhett
Net Consonance : consonant endings by net philanthropy, open source  
and other randomness