Blocking object creation for some users

Lists: pgsql-general
From: Bruno Wolff III <bruno(at)wolff(dot)to>
To: pgsql-general(at)postgresql(dot)org
Subject: Blocking object creation for some users
Date: 2002-04-01 21:33:07
Message-ID: [email protected]
Views: Whole Thread | Raw Message | Download mbox | Resend email
Lists: pgsql-general

Is there a way to keep users from creating new objects in a database
while letting them access existing objects in the same database?

It doesn't look like there is a clean (e.g. with GRANT) way to do this. I
tried looking to see if there was a sneaky way to block update access to some
of the system tables, but that doesn't seem to be doable either.


From: Steve Lane <slane(at)fmpro(dot)com>
To: <pgsql-general(at)postgresql(dot)org>
Subject: Postgres/PHP, Apache child processes dying
Date: 2002-04-03 16:10:08
Message-ID: B8D08580.B468%[email protected]
Views: Whole Thread | Raw Message | Download mbox | Resend email
Lists: pgsql-general

Hello all:

Been seeing a worsening problem on a number of my servers lately. I'm using
postgres as a backend to PHP applications, hosted on Apache. At one
particular point in the application (always in the web-side login, which is
probably significant), my web browser returns a "could not connect" error.

Scanning the logs, I find this in the postmaster log:

pq_recvbuf: unexpected EOF on client connection

And this in the Apache error log:

[Wed Feb 20 21:29:52 2002] [notice] child pid 18595 exit signal Segmentation
fault (11)

There seem to be a lot of reasons why child processes die in this way in
Apache. Does anyone know of any specific to postgres/PHP?

Thanks,

steve


From: "Thomas T(dot) Thai" <tom(at)minnesota(dot)com>
To: Steve Lane <slane(at)fmpro(dot)com>
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: Postgres/PHP, Apache child processes dying
Date: 2002-04-03 16:43:15
Message-ID: [email protected]
Views: Whole Thread | Raw Message | Download mbox | Resend email
Lists: pgsql-general

On Wed, 3 Apr 2002, Steve Lane wrote:

[...]
> pq_recvbuf: unexpected EOF on client connection
>
> And this in the Apache error log:
>
> [Wed Feb 20 21:29:52 2002] [notice] child pid 18595 exit signal Segmentation
> fault (11)

sounds like your apache process (probably with a DSO PHP) died
unexpectedly (any coredumps?) before it was suppose to complete the
transaction with PostgreSQL backend.

it could be anything that cause this. without a trace or further system
configuration, it would be hard to tell you what's going on.

--
Thomas T. Thai
Minnesota.com, Inc.


From: tony <tony(at)animaproductions(dot)com>
To: Steve Lane <slane(at)fmpro(dot)com>
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: Postgres/PHP, Apache child processes dying
Date: 2002-04-03 17:04:05
Message-ID: 1017853446.4998.5.camel@vaio
Views: Whole Thread | Raw Message | Download mbox | Resend email
Lists: pgsql-general

On Wed, 2002-04-03 at 18:10, Steve Lane wrote:

> pq_recvbuf: unexpected EOF on client connection
>
> And this in the Apache error log:
>
> [Wed Feb 20 21:29:52 2002] [notice] child pid 18595 exit signal Segmentation
> fault (11)
>
> There seem to be a lot of reasons why child processes die in this way in
> Apache. Does anyone know of any specific to postgres/PHP?

I have never seen this with JSP. Something wrong with your php code.

Cheers

Tony Grant

--
RedHat Linux on Sony Vaio C1XD/S
http://www.animaproductions.com/linux2.html
Macromedia UltraDev with PostgreSQL
http://www.animaproductions.com/ultra.html


From: Steve Lane <slane(at)fmpro(dot)com>
To: <pgsql-general(at)postgresql(dot)org>
Subject: Re: Postgres/PHP, Apache child processes dying
Date: 2002-04-03 20:03:21
Message-ID: B8D0BC29.B4B6%[email protected]
Views: Whole Thread | Raw Message | Download mbox | Resend email
Lists: pgsql-general

On 4/3/02 10:43 AM, "Thomas T. Thai" <tom(at)minnesota(dot)com> wrote:

> On Wed, 3 Apr 2002, Steve Lane wrote:
>
> [...]
>> pq_recvbuf: unexpected EOF on client connection
>>
>> And this in the Apache error log:
>>
>> [Wed Feb 20 21:29:52 2002] [notice] child pid 18595 exit signal Segmentation
>> fault (11)
>
> sounds like your apache process (probably with a DSO PHP) died
> unexpectedly (any coredumps?) before it was suppose to complete the
> transaction with PostgreSQL backend.

Yes, it's a DSO PHP. Eh, coredumps! Where would I look (betrays his helpless
ignorance).

-- steve