Authentication problem

Поиск
Список
Период
Сортировка
Искать

Authentication problem

От:
Dmitry Morozovsky <marck@rinet.ru>
Дата:
Hi there colleagues,

From the Docs (Admin 4.1):

	There is no "fall-through" or "backup": if one record is chosen
	and the authentication fails, the following records are not
	considered.

Are there any plans to loose this restriction? It would be very useful
to use e.g.

local	all		ident		admin
local	sameuser	ident		sameuser
local	all		password	passwd.user


and have backup pseudo-users in admin ident-map, allow connecting users to
personal databases and list exceptions in password file.

Or, is there another way to achieve this?

Also, of course, it would be _very_ useful to tell full connects and
read-only connects (not allowed to create tables/indexes/views/etc...)

Sincerely,
D.Marck                                   [DM5020, DM268-RIPE, DM3-RIPN]
------------------------------------------------------------------------
*** Dmitry Morozovsky --- D.Marck --- Wild Woozle --- marck@rinet.ru ***
------------------------------------------------------------------------

Re: Authentication problem

От:
Tom Lane <tgl@sss.pgh.pa.us>
Дата:
Dmitry Morozovsky  writes:
> 	There is no "fall-through" or "backup": if one record is chosen
> 	and the authentication fails, the following records are not
> 	considered.

> Are there any plans to loose this restriction?

No.  I don't believe we could count on clients to respond to multiple
authentication challenges of different types.

> It would be very useful to use e.g.

> local	all		ident		admin
> local	sameuser	ident		sameuser
> local	all		password	passwd.user

The "sameuser" part of this works now, since sameuser is a record
matching constraint, not an authentication test.

There has been some talk of adding a more flexible username-matching
field to pg_hba (whereupon the file name would be inappropriate ;-))
but no one's really done any work on it.

			regards, tom lane

Re: Authentication problem

От:
Bruce Momjian <pgman@candle.pha.pa.us>
Дата:
> The "sameuser" part of this works now, since sameuser is a record
> matching constraint, not an authentication test.
> 
> There has been some talk of adding a more flexible username-matching
> field to pg_hba (whereupon the file name would be inappropriate ;-))
> but no one's really done any work on it.

I hope to add the username for 7.3.

-- 
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026
FAQ