Re: ident changes between 8.3 and 8.4 - Mailing list pgsql-hackers

From Andrew Dunstan
Subject Re: ident changes between 8.3 and 8.4
Date
Msg-id 4AF2E331.3010102@dunslane.net
Whole thread Raw
In response to Re: ident changes between 8.3 and 8.4  (Magnus Hagander <magnus@hagander.net>)
Responses Re: ident changes between 8.3 and 8.4
Re: ident changes between 8.3 and 8.4
List pgsql-hackers

Magnus Hagander wrote:
> On Thu, Nov 5, 2009 at 14:49, Rafael Martinez <r.m.guerrero@usit.uio.no> wrote:
>   
>> Hash: SHA1
>>
>> Hello
>>
>> We have been using for years and without problems local ident
>> autentification in the database for our user postgres.
>>
>> These are the values that we have been using until version 8.3:
>> pg_hba.conf:
>> - ------------
>> local    all         postgres       ident sameuser
>>
>> pg_ident.conf:
>> - --------------
>> sameuser         postgres          postgres
>>
>> With 8.4, we get this error if we use a map named 'sameuser'.
>> - -----------------------------------------------------------
>> FATAL:  Ident authentication failed for user "postgres"
>> LOG:  no match in usermap for user "postgres" authenticated as "postgres"
>> CONTEXT:  usermap "sameuser"
>> - -----------------------------------------------------------
>>
>> These are the values used with 8.4:
>> pg_hba.conf:
>> - ------------
>> local    all         postgres       ident map=sameuser
>>
>> pg_ident.conf:
>> - --------------
>> sameuser         postgres          postgres
>>
>> After some investigation, we have found out that everything works
>> without problems if we change the mapname used by ident to something
>> different than 'sameuser'.
>>
>> Is this a bug or have we decided this behavior? I can not find any
>> documentation explaining that 'sameuser' is not a valid mapname.
>>     
>
> To make 8.4 behave like the previous "ident sameuser" way, just put
> "ident". No map is needed.
>   


And it is documented in the release notes: 
<http://www.postgresql.org/docs/current/static/release-8-4.html>, which 
the OP should have read when upgrading.

cheers

andrew


pgsql-hackers by date:

Previous
From: Magnus Hagander
Date:
Subject: Re: ident changes between 8.3 and 8.4
Next
From: Robert Haas
Date:
Subject: Re: operator exclusion constraints