More on Compress/Uncompress

Only very lately I figured out that there is a way to compress/uncompress files to a different directory / location/ folder on UNIX.  This is very important while taking a compressed OS backup or restoring compressed files from backup.

To compress to a different folder / locaiton :

compress -c /appl01/orabkup/system01.dbf > /u01/oradata/ORCL/ORCL_system01.dbf.Z
or
compress < /appl01/orabkup/system01.dbf > /u01/oradata/ORCL/ORCL_system01.dbf.Z

To uncompress to a different folder / locaiton :

uncompress -c /appl01/orabkup/ORCL_system01.dbf.Z > /u01/oradata/ORCL/system01.dbf
or
uncompress < /appl01/orabkup/ORCL_system01.dbf.Z > /u01/oradata/ORCL/system01.dbf

This may be a very primitive one, but I learned it very recently. Hope it helps you!

10g Upgrade – issues with DBMS_STATS package

Last week, one of our clients production database was upgraded from 9i to 10g.  Since then they were facing lot of performance problems.  They cannot pin point the problem at a particular place, so the problem seems to be “overall”.  I did not have access to the production box, all I had was an AWR report generated during the time of performance degradation.  When I dug into the report, I found that the instance performance is generally ok.  The problem was with most of the application queries – too many physical reads.

I remember I read an article from Jonathan Lewis on his blog regarding performance degradation after 10g upgrade.  I quickly went there and did a search and found out this URL http://jonathanlewis.wordpress.com/2007/02/02/10g-upgrade/, excellently explaining the change in the behaviour of dbms_stats package.  In short.

“If you didn’t specify a method_opt in your scripts under 9i you were not generating histograms; but under 10g you will be collecting histograms on any columns that Oracle thinks might be suitable candidates.” 

I then checked the AWR report to see if there was anything recorded regarding DBMS_STATS, indeed I found out that they were using defualt method_opt

dbms_stats.gather_table_stats(
OwnName => 'ADMIN',tabname=>'T2838',estimate_percent => dbms_stats.auto_sample_size,
Granularity => 'DEFAULT',Degree => 4)

Which means they were generating histograms, but on 9i the same script would not have generated them.  This could be having a major impact on generating execution plans on 10g.  I replied them this, providing the Jonathan Lewis’ URL.

Later, they did reset the METHOD_OPT to 9i levels.

exec dbms_stats.set_param('METHOD_OPT', 'FOR ALL COLUMNS SIZE 1');

I am waiting to hear from them on whether there is any improvement in performance.

Note:  I later found Richard Foote has also written an excellent article on this : http://richardfoote.wordpress.com/2008/01/04/dbms_stats-method_opt-default-behaviour-changed-in-10g-be-careful/

Database refresh using cold backup

Sometime, you will be asked to refresh an UAT database using an coldbackup of production one.  Here are the steps :

1.  Prepare controlfile creation script.  This can be done by

alter database backup controlfile to trace as '/export/home/oracle/ctlfile.sql';

2.  View the edit ctlfile.sql, so that it should contain the following

CREATE CONTROLFILE SET DATABASE "UATDB" RESETLOGS NOARCHIVELOG
  MAXLOGFILES 16
  MAXLOGMEMBERS 3
  MAXDATAFILES 100
  MAXINSTANCES 1
  MAXLOGHISTORY 453
LOGFILE
  GROUP 1 '/u01/oradata/UATDB/redo01.log' SIZE 200M,
  GROUP 2 '/u02/oradata/UATDB/redo02.log' SIZE 200M,
  GROUP 3 '/u03/oradata/UATDB/redo03.log' SIZE 200M
-- STANDBY LOGFILE
DATAFILE
  '/u01/oradata/UATDB/system01.dbf',
  '/u01/oradata/UATDB/undotbs01.dbf',
  '/u01/oradata/UATDB/users01.dbf',
  '/u01/oradata/UATDB/data01.dbf',
  '/u01/oradata/UATDB/tools01.dbf',
  '/u01/oradata/UATDB/perfstat01.dbf',
  '/u01/oradata/UATDB/sysaux01.dbf'
CHARACTER SET WE8ISO8859P1
;

3. Gernerate script to create TEMP tablespace

set long 2000
select DBMS_METADATA.GET_DDL('TABLESPACE','TEMP') from dual;

4. Shutdown UATDB cleanly

shutdown immediate

5. Copy the cold backup files from production to UAT box.

6. Start the database in nomount state
 
startup nomount

7. Run the create control file script

@/export/home/oracle/ctlfile.sql

8. Open the database with resetlogs

alter database open resetlogs;

9. Create temp tablespace from the script generated from step 3:

CREATE TEMPORARY TABLESPACE "TEMP1" TEMPFILE
'/u01/oradata/UATDB/TEMP01.dbf' SIZE 5242880000
EXTENT MANAGEMENT LOCAL UNIFORM SIZE 1048576

Extracting one or more files from compressed tar archive

As a oracle DBA you will come across extracting files from compressed tar archive files like *.gz/*.Z normally compressed through gzip or UNIX compress utility.  You can uncompress them and extract the tar file.  But it is a two step process and you need more diskspace to do this (twice the amount nearly) .  Instead you can do them in one step. 

How to do this with gzip is clearly demonstrated in here in this URL http://www.sron.rug.nl/didac/didac_tips.html.

gzip -dc t.tar.gz | tar -tvf -

That saves you lot of disk space. 

 Now lets try the same with UNIX uncompress utility.

I have 3 text files and I add them to t.tar
$ ls -ltr
total 146
-rw-r--r-- 1 oracle dba 6 Jan 26 05:28 t1.txt
-rw-r--r-- 1 oracle dba 6 Jan 26 05:28 t2.txt
-rw-r--r-- 1 oracle dba 6 Jan 26 05:28 t3.txt
$ tar -cvf t.tar *.txt
a t1.txt 1K
a t2.txt 1K
a t3.txt 1K


Now compress them using UNIX compress command

$ compress t.tar
$ ls -ltr
total 148
-rw-r--r-- 1 oracle dba 296 Jan 26 05:31 t.tar.Z


Try the uncompress command

$ uncompress -c t.tar.Z | tar -tvf -
tar: blocksize = 8
-rw-r--r-- 101/201 6 Jan 26 05:28 2008 t1.txt
-rw-r--r-- 101/201 6 Jan 26 05:28 2008 t2.txt
-rw-r--r-- 101/201 6 Jan 26 05:28 2008 t3.txt


Yes!! it really works!  If you want to extract just one file, you can do the below

$ uncompress -c t.tar.Z | tar -xvf - t1.txt
tar: blocksize = 8
x t1.txt, 6 bytes, 1 tape blocks
$ ls -ltr
total 144
-rw-r--r-- 1 oracle dba 6 Jan 26 05:28 t1.txt
-rw-r--r-- 1 oracle dba 296 Jan 26 05:31 t.tar.Z


Configuring wireless network

I have my ADSL modem and PC on the upstairs of my home. It is good, because I am isolated and work in peace when needed.  But carrying laptop to upstairs can sometime be tough, especially you need to work for a short duration.  So I thought of configuring wireless network for this purpose. Yesterday I did configure my HUAWEI modem WA1003A – BSNL broad band to work wireless.  Just thought I would give it in steps so that it can help someone.  

Router/Modem configuration:

1.  Router type WA1003A HUAWEI.  Brodband vendor – BSNL Chennai
2.  Log on to the admin website from a PC that is connected to network.  Usually it will be http://192.168.1.1.
3.  On the top menu, click “Setup”.  On the left-hand side menu click “Wireless”. 
4.  Check “Enable AP”. Give a name for SSID.  The SSID uniquely identifies your wireless network. Click “Apply”
5.  On the top menu, click “Advanced”.  On the left-hand side menu click “Wireless Security”.
6.  Select the kind of security you wish to have.  Though this is optional, it is absolutely necessary to have this.
    You don’t want an intruder connected to your router.
7.  Check “WEP” option.
8.  Check “Enable WEP Wireless Security”
9.  Select Authentication Type “Shared”
10. Give an encryption key in hexa decimal.  Select cipher as “64 bits”
11. Apply the changes.
12. Go to Tool (top menu) -> System commands (side menu) -> Save all.  After saving, click restart.
13. Now your router will be restarted.  The wireless LED on the modem should be on now.

Laptop Configuration:

1.  Enable wireless.  On my Dell D610 laptop it is the combination of “Fn” + “F2″ keys.
2.  Go to Start -> My Network Places.
3.  Right click “Wireless Network Connection” -> View available wireless networks.
4.  You should find your wireless network there. Select it and click connect.
5.  You will be asked to enter the key.  Enter the hexa decimal key you have set while configuring the modem.
6.  That’s it, you are done!!! I always use DHCP, so no need to bother about IP addresses.

Hope this helps you!!

It served the purpose..

Couple of months back, I was asked to run a script on UAT as part of a change deployment.  When I looked into it, I got shocked.  You can see yourself why..

REM   
REM    NAME
REM    abc.sql - Delete Contact
REM
REM       ********************************************************
PROMPT    *** abc.sql ...                                      ***
set define off
set serveroutput on
delete from XXXXXXXX where EMAIL='abc.xys@oracle.com';
dbms_output.put_line('*** Script Confirmation : Property Deleted from XXXXXXXX ***');
commit;
exception
  when others then
    dbms_output.put_line('*** Error: sqlcode: ' || sqlcode || '; sqlerrm: ' || substr(sqlerrm, 1, 100) || ' ***');
PROMPT    *** abc.sql ... completed                            ***
REM       ********************************************************
EXIT

If you notice, there is no DECLARE block, no BEGIN block, only exception block!!!!
I asked the developer how it didn’t give errors on development environment? He told “yes, but the record got deleted and commit was done.  That’s all I want.  It served the purpose.”
He was one among those who attended training on PL/SQL earlier that month.  Needless to say, he has learnt the “Error handling” very well, at least.

OEL on VMWare

Finally sometime back, I have completed installing Oracle Enterprise Linux on VMWare server.  The installation is quite simple and straight forward.  I just followed the steps given by Frederik (Frikkie) Visser in http://startoracle.com/2007/09/30/so-you-want-to-play-with-oracle-11gs-rac-heres-how/ .  It is an excellent article that tells you step by step not only how to install linux on VMWare but also configuring RAC.  A must read for a DBA. Thanks Frikkie.

My first reaction to OEL is, it is bulky and a bit slow.  But I have allocated only 800 MB or memory to it.  May be I should throw more memory to it and see how it behaves.

My next step would be to download Oracle 11g for Linux ans start installing it.

VMServer 2.0 Beta 1

I downloaded VMServer 2.0 Beta 1 today from here and installed it.   It has a web interface for creating and administering VM images.  It uses Tomcat webserver for this purpose. 

For several minutes I struggled to log in to the web console, because it is asking for username and password.  It never asked me these details while installing.  The username and passwords on tomcat-users.xml didn’t work.  Finally in the release note, I figured it out, that it uses OS authentication. 

But, again there is a small issue for me.  Since I only use my PC, i do not have passwords for the OS accounts.  But the VMWare web console doesn’t take an empty password. hmmmmmmm…… quite frustrating..  Finally I gave up and set password for my Windows OS account… Grrrrrrrr…

Finally, I could able to get into VMWare console.

Nice X-Window emulator

On Windows, at-least, once in a while DBAs need to remotely access the UNIX/Linux GUI tools such as OUI, DBCA etc from putty or an SSH client.   Tools like Exceed are very expensive and VNC port forwarding is a bit of complex solution (also you need vnc software).  

To overcome this, you can try this tiny little software called Xming.  The setup file is only about 2MB of size and very easy to install.  All you need to do is to start the Xming, open a putty session and check “Enable X11 Forwarding” option under SSH->X11.   Log into the Linux/Unix sever and straight away you can test it by running xlogo/xclock.  No need to even set the DISPLAY env variable.  Very nice indeed.

I just tried it and it is absolutely working fine .

Standby v$archived_log

It happened last week, that one of my colleague came to me in a hurry and said that the archived logs are not getting applied on standby (Oracle 10gR1).  When I went and checked the standby database, everything was OK.  It applied till the last archived log.  On the primary database side though, the status was ‘NO’ for few of the archive logs

Standby:

SQL> SELECT SEQUENCE#, ARCHIVED, APPLIED FROM V$ARCHIVED_LOG WHERE SEQUENCE# > 11210 ORDER BY SEQUENCE#;

SEQUENCE# DEST_ID ARC APP
———- ———- —
11211 YES YES
11212 YES YES
11213 YES YES
11214 YES YES
11215 YES YES
11216 YES YES
11217 YES YES
11218 YES YES
11219 YES YES

Primary:

SQL> SELECT SEQUENCE#, ARCHIVED, APPLIED FROM V$ARCHIVED_LOG WHERE DEST_ID = 2 AND SEQUENCE# > 11210 ORDER BY SEQUENCE#;

SEQUENCE# DEST_ID ARC APP
———- ———- —
11211  YES YES
11212  YES YES
11213  YES NO
11214  YES YES
11215  YES NO
11216  YES YES
11217  YES YES
11218  YES NO
11219  YES YES

 I did a search on metalink and came across this document ‘Note:263994.1‘.  This is what mentioned as cause :

“The cause of the problem is that the APPLIED column of V$ARCHIVED_LOG on the primary is not always updated. If the sequence number you look at was actually sent as a GAP (could not be sent originally and was later sent by an ARCH process as part of a FAL request).

In such a case the APPLIED column for that sequence number will not be updated. “

The conclusion is that, we cannot fully trust the v$archived_log on primary server.