X-Original-To: pgsql-admin-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [200.46.204.144]) by svr1.postgresql.org (Postfix) with ESMTP id 433065E46F4 for ; Thu, 26 Aug 2004 19:17:12 -0300 (ADT) Received: from svr1.postgresql.org ([200.46.204.71]) by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024) with ESMTP id 40579-09 for ; Thu, 26 Aug 2004 22:17:06 +0000 (GMT) Received: from web61308.mail.yahoo.com (web61308.mail.yahoo.com [216.155.196.151]) by svr1.postgresql.org (Postfix) with SMTP id F2A605E46EC for ; Thu, 26 Aug 2004 19:17:04 -0300 (ADT) Message-ID: <20040826211532.30654.qmail@web61308.mail.yahoo.com> Received: from [68.101.43.86] by web61308.mail.yahoo.com via HTTP; Thu, 26 Aug 2004 14:15:32 PDT Date: Thu, 26 Aug 2004 14:15:32 -0700 (PDT) From: kris pal Subject: Re: regression database To: kris pal , Tom Lane Cc: pgsql-admin@postgresql.org In-Reply-To: <20040826173209.18246.qmail@web61306.mail.yahoo.com> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="0-1082699247-1093554932=:28430" X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, hits=0.0 tagged_above=0.0 required=5.0 tests=HTML_MESSAGE X-Spam-Level: X-Archive-Number: 200408/269 X-Sequence-Number: 14664 --0-1082699247-1093554932=:28430 Content-Type: text/plain; charset=us-ascii Actually I just noticed something. Previously for some reason, I thought the data directory corresponded to 'regression'. But I double checked it now with oid2name and it corresponds to our main database say "dds". But the issue still remains. In the sense that pg_attribute etc are still so huge and I have no idea why. This table is not the one we created in the schema. thanks, Kris Pal. kris pal wrote:Hi tom, It is actually named 'regression'. I spoke with the person who build the server but apparently he is not well worsed in Postgres. He didn't explicitly create 'regression', I guess he selected some default installation options. I am not sure how to find out what is causing the 'regression' database to increase in size. Its size is not due to the tables with in the 'regression' but due to the system tables like "pg_attribute" with in 'regression' database. If you chould point to a resource which would help me understand how this 'regression' database is handled internally (what activity writes to this database) that would be great. Pls see below, the directory 2034862 corresponds to 'regression': ____________________________________________________ [postgres@dds base]$ du -sk /home/postgres/data/base/2034862 6236920 /home/postgres/data/base/2034862 [postgres@dds base]$ [postgres@dds base]$ cd /home/postgres/data/base/2034862 [postgres@dds 2034862]$ du -sk * | awk '{if ($1 > 512000) print $1" --> "$2}' 1049604 --> 1249 (over 1GB) 949484 --> 1249.1 (abt 1GB) 1049604 --> 16609 821104 --> 16609.1 The relname for the corresponding relfilenode: --------------+------------- pg_attribute | 1249 pg_attribute_relid_attnam_index | 16609 --------------+------------- ____________________________________________________ So am not sure why this thing is growing progressively. Any feedback/ help is greatly appreciated. thanks, Kris Pal. Tom Lane wrote: kris pal writes: > I am not sure what the implications of completely deleting the > 'Regression' database (I don't need the regression testing). Is it named "Regression" with a capital R? If so, that's not the DB that is made by the standard regression tests --- that is called "regression". There is not any reason not to drop the standard regression database, if you're not using it. But there is no way that that DB would get that large unless something in your installation is using it. I'd suggest you find out what. regards, tom lane --------------------------------- Do you Yahoo!? New and Improved Yahoo! Mail - Send 10MB messages! --------------------------------- Do you Yahoo!? Win 1 of 4,000 free domain names from Yahoo! Enter now. --0-1082699247-1093554932=:28430 Content-Type: text/html; charset=us-ascii
 
Actually I just noticed something. Previously for some reason, I thought the data directory corresponded to 'regression'. But I double checked it now with oid2name and it corresponds to our main database say "dds".
 
But the issue still remains. In the sense that pg_attribute etc are still so huge and I have no idea why. This table is not the one we created in the schema.
 
thanks,
Kris Pal.


kris pal <kristhegambler@yahoo.com> wrote:
Hi tom,
 
It is actually named 'regression'. I spoke with the person who build the server but apparently he is not well worsed in Postgres. He didn't explicitly create 'regression', I guess he selected some default installation options.
 
I am not sure how to find out what is causing the 'regression' database to increase in size. Its size is not due to the tables with in the 'regression' but due to the system tables like "pg_attribute" with in 'regression' database. If you chould point to a resource which would help me understand how this 'regression' database is handled internally (what activity writes to this database) that would be great.
 
 
Pls see below, the directory 2034862 corresponds to 'regression':
 
____________________________________________________
 
[postgres@dds base]$ du -sk /home/postgres/data/base/2034862
6236920 /home/postgres/data/base/2034862
[postgres@dds base]$
[postgres@dds base]$ cd /home/postgres/data/base/2034862
[postgres@dds 2034862]$ du -sk * | awk '{if ($1 > 512000)  print $1" --> "$2}'
1049604 --> 1249    (over 1GB)
949484 --> 1249.1   (abt 1GB)
1049604 --> 16609
821104 --> 16609.1
The relname for the corresponding relfilenode:
--------------+-------------
 pg_attribute |        1249
 pg_attribute_relid_attnam_index |       16609
--------------+-------------
____________________________________________________
 
 
So am not sure why this thing is growing progressively.
Any feedback/ help is greatly appreciated.
 
thanks,
Kris Pal.


Tom Lane <tgl@sss.pgh.pa.us> wrote:
kris pal writes:
> I am not sure what the implications of completely deleting the
> 'Regression' database (I don't need the regression testing).

Is it named "Regression" with a capital R? If so, that's not the DB that
is made by the standard regression tests --- that is called "regression".

There is not any reason not to drop the standard regression database, if
you're not using it. But there is no way that that DB would get that
large unless something in your installation is using it. I'd suggest
you find out what.

regards, tom lane


Do you Yahoo!?
New and Improved Yahoo! Mail - Send 10MB messages!


Do you Yahoo!?
Win 1 of 4,000 free domain names from Yahoo! Enter now. --0-1082699247-1093554932=:28430--