Re: Initdb on Windows 2003

Поиск
Список
Период
Сортировка
От
Тема Re: Initdb on Windows 2003
Дата
Msg-id 2E22904642859A40AF5E1CA1597EBB1F02EF4244@devexmb1.corp.nai.org
обсуждение исходный текст
Ответ на Initdb on Windows 2003  (<James_Hughes@McAfee.com>)
Список pgsql-hackers
All

I have determined what is causing the failure. It appears that the
stdout & stderr redirection to nul produces the "Access is Denied."
message. This is happening even if I type "dir >nul" at the command
prompt! I assume that this re-direction in PostgreSQL is done when
starting postgres.exe as to allow the communction between the two
processes?

I know this is not the correct place to ask my next question but maybe
there is a Windows expert out there :o) Does anyone have any idea how
Windows controls access to devices such as nul,comN, lpt1, etc? Or what
security setting governs this?

Thanks all for your help.

Regards
James

-----Original Message-----
From: pgsql-hackers-owner@postgresql.org
[mailto:pgsql-hackers-owner@postgresql.org] On Behalf Of Hughes, James
Sent: 02 March 2006 07:51
To: andrew@dunslane.net
Cc: pgsql-hackers@postgresql.org
Subject: Re: [HACKERS] Initdb on Windows 2003

Thanks for the info, I shall indeed try this.

One thing to mention is that I don't think the error occurs within
initdb, it seams to be postgres.exe that dbinit starts.

Regards
James

-----Original Message-----
From: Andrew Dunstan [mailto:andrew@dunslane.net]
Sent: 01 March 2006 16:26
To: Hughes, James
Cc: pgsql-hackers@postgresql.org
Subject: Re: [HACKERS] Initdb on Windows 2003


(private email says leaving off --user=foo doesn't fix it ;-( )

Since you have apparently compiled your own, could you please try with
the latest stable initdb.c code for your release? That is version
1.99.2.2 for release 8.1 or 1.73.4.3 for release 8.0. They are
downloadable here:
http://developer.postgresql.org/cvsweb.cgi/pgsql/src/bin/initdb/initdb.c

Maybe there's something odd about your setup - we have a WS2k3 machine
happily building and running on buildfarm: see
http://www.pgbuildfarm.org/cgi-bin/show_history.pl?nm=snake&br=REL8_1_ST
ABLE

If it still doesn't work, you might get some useful info from initdb
--debug

Thanks

andrew


James_Hughes@McAfee.com wrote:

>Hi,
>
>1. The command line passed is -D "c:\data" --user=McAfeePostgresUser 2.

>McAfeePostgresUser which is a local machine user with limited
>privileges. Though this problem occurs with any user account you create

>with non-administrator privileges.
>
>Thanks
>James
>
>-----Original Message-----
>From: Andrew Dunstan [mailto:andrew@dunslane.net]
>Sent: 01 March 2006 13:27
>To: Hughes, James
>Cc: pgsql-hackers@postgresql.org
>Subject: Re: [HACKERS] Initdb on Windows 2003
>
>
>
>1. please show the EXACT initdb command line used.
>
>2. Which Windows user was actually running initdb?
>
>cheers
>
>andrew
>
>
>
>>Hello all,
>>
>>Below is an email thread regarding a possible bug in PostgreSQL on
>>Windows 2003. Any help or advice anyone can give on this would be much
>>
>>
>
>
>
>>appreciated.
>>
>>All the best,
>>James Hughes
>>
>>--- EMAIL THREAD FOLLOWS ---
>>
>>James_Hughes@McAfee.com wrote:
>>
>>
>>>Hi Bruce,
>>>
>>>We are now seeing this issue on three machines, all of which are
>>>running Windows 2003. After some looking at the code and putting some
>>>
>>>
>
>
>
>>>extra debug output (very little) all we have determined is that
>>>initdb.exe opens postgres.exe via pipes, and at some point within
>>>postgres.exe this error is generated.
>>>
>>>If we knock out the check for the user being not being admin and run
>>>initdb.exe as an administrator all works as expected. Adding the user
>>>
>>>
>
>
>
>>>to the 'Power Users' group also causes the "Access Denined" error! I
>>>don't believe that the access denined is anything to do with file
>>>access or memory (shared) access as I ran some tools to see what
>>>postgres.exe is trying to access.
>>>
>>>Is there an easy way on Windows to step-through the PostgreSQL code
>>>as
>>>
>>>
>>>it runs? Is there any other information I can provide which will help
>>>
>>>
>
>
>
>>>you with this defect?
>>>
>>>Any help getting to the root of this problem is much appericated.
>>>
>>>All the best
>>>James Hughes
>>>
>>>-----Original Message-----
>>>From: Bruce Momjian [mailto:pgman@candle.pha.pa.us]
>>>Sent: 24 February 2006 18:54
>>>To: Hughes, James
>>>Cc: pgsql-bugs@postgresql.org
>>>Subject: Re: [BUGS] BUG #2268: initdb fails to initialize database
>>>cluster with error "Access is denied"
>>>
>>>
>>>Strange.  It isn't a typical error we see, and the fact you see it on
>>>
>>>
>
>
>
>>>two machines is even stranger.  My guess is that somehow the
>>>configuration on those two machines is the same and is causing the
>>>failure.
>>>
>>>---------------------------------------------------------------------
>>>-
>>>--
>>>---
>>>
>>>James Hughes wrote:
>>>
>>>
>>>>The following bug has been logged online:
>>>>
>>>>Bug reference:      2268
>>>>Logged by:          James Hughes
>>>>Email address:      james_hughes@mcafee.com
>>>>PostgreSQL version: 8.1.3
>>>>Operating system:   Windows 2003 (Server)
>>>>Description:        initdb fails to initialize database cluster
>>>>
>>>>
>with
>
>
>>>error
>>>
>>>
>>>>"Access is denied"
>>>>Details:
>>>>
>>>>When we try and initialize a database cluster using initdb.exe we
>>>>are getting the following output.
>>>>
>>>>-- OUTPUT from dbinit.exe --
>>>>The files belonging to this database system will be owned by user
>>>>
>>>>
>>>"user1".
>>>
>>>
>>>>This user must also own the server process.
>>>>
>>>>The database cluster will be initialized with locale English_United
>>>>
>>>>
>
>
>
>>>>Kingdom.1252.
>>>>
>>>>creating directory c:/dataa ... ok
>>>>creating directory c:/dataa/global ... ok creating directory
>>>>c:/dataa/pg_xlog ... ok creating directory
>>>>c:/dataa/pg_xlog/archive_status ... ok creating directory
>>>>c:/dataa/pg_clog ... ok creating directory c:/dataa/pg_subtrans ...
>>>>ok
>>>>
>>>>
>>>>creating directory c:/dataa/pg_twophase ... ok creating directory
>>>>c:/dataa/pg_multixact/members ... ok creating directory
>>>>c:/dataa/pg_multixact/offsets ... ok creating directory
>>>>c:/dataa/base ... ok creating directory c:/dataa/base/1 ... ok
>>>>creating directory c:/dataa/pg_tblspc ... ok selecting default
>>>>max_connections ... Access
>>>>
>>>>
>>>>is denied.
>>>>Access is denied.
>>>>Access is denied.
>>>>Access is denied.
>>>>Access is denied.
>>>>Access is denied.
>>>>10
>>>>selecting default shared_buffers ... Access is denied.
>>>>Access is denied.
>>>>Access is denied.
>>>>Access is denied.
>>>>Access is denied.
>>>>Access is denied.
>>>>Access is denied.
>>>>Access is denied.
>>>>Access is denied.
>>>>Access is denied.
>>>>Access is denied.
>>>>50
>>>>creating configuration files ... ok creating template1 database in
>>>>c:/dataa/base/1 ... ok initializing pg_authid ... Access is denied.
>>>>child process was terminated by signal 1
>>>>initdb: removing data directory "c:/dataa"
>>>>
>>>>-- end: OUTPUT from dbinit.exe --
>>>>
>>>>user1 is a user without administrative rights. I have checked all
>>>>the directories and the user has permissons to write to the data
>>>>directory, and read other files it requires.
>>>>
>>>>I also tried the Windows installer (MSI) from pgFoundry. This too is
>>>>
>>>>
>>>>having the same problem.
>>>>
>>>>We have two machines which this problem occurs on, one of which
>>>>originally the initialization would work on. I have checked all the
>>>>
>>>>
>
>
>
>>>>user rights asssignments, and even given the postgres user (user1
>>>>in
>>>>
>>>>
>>>>the above output from dbinit.exe) all user rights.
>>>>
>>>>---------------------------(end of
>>>>broadcast)---------------------------
>>>>TIP 2: Don't 'kill -9' the postmaster
>>>>
>>>>
>>>>
>>>--
>>>  Bruce Momjian   http://candle.pha.pa.us
>>>  SRA OSS, Inc.   http://www.sraoss.com
>>>
>>>  + If your life is a hard drive, Christ can be your backup. +
>>>
>>>
>>>
>>--
>> Bruce Momjian   http://candle.pha.pa.us
>> SRA OSS, Inc.   http://www.sraoss.com
>>
>> + If your life is a hard drive, Christ can be your backup. +
>>
>>
>>
>>
>>
>>
>>James Hughes
>>
>>
>>---------------------------(end of
>>broadcast)--------------------------- TIP 6: explain analyze is your
>>friend
>>
>>
>
>
>
>
>


---------------------------(end of broadcast)---------------------------
TIP 5: don't forget to increase your free space map settings


В списке pgsql-hackers по дате отправления:

Предыдущее
От: Andrew - Supernews
Дата:
Сообщение: Re: ACCESS EXCLUSIVE LOCK
Следующее
От: Martijn van Oosterhout
Дата:
Сообщение: Re: Automatic free space map filling