zaterdag 24 augustus 2013

Troubleshooting & Solving Linux disk space problems

In this blogpost I want to explain how you can troubleshoot diskspace problems.

Find out which files are using all the disk space.

Graphical desktop

When you have a graphical desktop at hand you should really consider installing Baobab (also known as disk usage analyzer for Ubuntu users). This application is rather straightforward.

You can click on 'Scan home' to verify the disk usage in your home directory. You can also click the button with the Harddisk symbol to start the analysis at your root directory. The graph on the right shows the usage per folder, this is handy to quickly identify greedy disk space consumers.

Command line

When you don't have a graphical desktop at hand, you must help yourself with the df and du commands.


The df tool will show you a table that reports the file system disk space usage. It takes a -h flag to write the output in human readable format. A table (constructed by df -h) looks like the following:


As you can see it is a handy tool to quickly see the state of your filesystems. A good first indication to see if problems are at hand.

The du tool will estimate the disk space usage. It takes 2 very handy flags which I cannot live without anymore.

  • -h -> Human readable
  • --max-depth=N -> Make a summary of directories up till the Nth level

The human readable flag will convert the numerical output from bytes (B) to more readable units like KB, MB, GB. This is the same behavior as with the df command.

Let's illustrated the du command with an example::

ow@D ~/Desktop $ du -h --max-depth=1 /tmp
du: cannot read directory `/tmp/vmware-root': Permission denied
4.0K /tmp/vmware-root
8.0K /tmp/pulse-O0P9w3N0W4t6
120G /tmp/img
8.0K /tmp/mintUpdate
du: cannot read directory `/tmp/cvcd': Permission denied
4.0K /tmp/cvcd
4.0K /tmp/ssh-fyPBrSQD2054
4.0K /tmp/orbit-peter
8.0K /tmp/sni-qt_skype_2251-QXMSXq
4.0K /tmp/.ICE-unix
4.0K /tmp/VMwareDnD
8.0K /tmp/matecorba-peter
4.0K /tmp/.X11-unix
4.0K /tmp/
4.0K /tmp/hsperfdata_peter
4.0K /tmp/jna
4.0K /tmp/keyring-GJe9xq
du: cannot read directory `/tmp/pulse-PKdhtXMmr18n': Permission denied
4.0K /tmp/pulse-PKdhtXMmr18n
164K /tmp

This example shows some typical output lines.

Each line represents a directory and the first column shows the size of the content of the directory.

The last line is a summary line and shows the total size of the content of /tmp (or at least an estimate of it taking into account the files/directories that the user can read).

The lines ending with permission denied illustrated that the du tool couldn't descend in these directories. This is because the user that launched the du command didn't have the rights to descend into that directory.

This usage can be used to track down the big disk space consumers. I prefer to start examining the disk usage with max-depth=1 from the root directory and go down the directories to find the files that consume too much disk space. E.g. Using df I notice that my filesystem mounted at /tmp is reaching it's limit (it has only 121GB of disk space and it consumes more then 120GB). Therefore I launch the du command from above and I get the output from the example above. Then I know that the disk space is used inside /tmp/img so I could launch du --max-depth=1 -h /tmp/img to drill down further. These steps can be repeated until you find the consuming files.

Mismatch between du & df.

There can be a mismatch between the outputs of du and df. This is because df really shows the diskspace that is free and du calculates the disk usage based on the file attributes. It can however be that a file you have deleted isn't really gone yet and thus takes up space. Eventhough it isn't completely gone, it doesn't show up in the directory listing and thus du won't take it into account.

An example of a situation where such a mismatch might occur is when you have an application that logs to a file. When you delete the file it can be that the application still had a handle to the file. The file won't show up in the directory listing anymore but the filespace is not freed up until the file handle is released. If this was a log file from an application that has high availability restrictions, this can be problematic since you cannot shut it down whenever you want. It is possible to force Linux to give away the space but it involves actions with /proc. This is something I might post about on a later time.

vrijdag 8 februari 2013

Use Oracle Database Gateway for ODBC to query MSSQL Server from an Oracle Database.


In this blogpost I will explain how you can setup Oracle Database Gateway for ODBC (=DG4ODBC) on a linux host to connect from an Oracle Database to a Microsoft SQL Server. To achieve this we will setup the unixODBC driver manager and use the freetds driver.

Overview of the solution.

The solution


Working databases

It is not the purpose of this blogpost to tell you how to install an Oracle or MS SQL Database. We will therefore only explain the setup of the DG4ODBC.

A (virtual) machine on which you can install DG4ODBC.

A (virtual) machine with a linux OS and a 64 bit architecture is required. I used an Oracle Linux 6.3 (64-bit) OS but any Red Hat (derivative) should suffice. The installation of the OS is rather basic. On the hard disk I only created 1 EXT4 partition and 1 swap partion. I chose 1 CPU with 2 cores and 2GiB of RAM.

Installed packages

I just did a server installation (no desktop environment). After the installation I installed the oracle-rdbms-server-11gR2-preinstall package (see for more information). This package is installed because I use this VM as a template to create test machines. It is not a requirement for the setup.

Actually if you want to install the package you can execute the yum statement below. Besides installing the packages the package also sets OS parameter to tune the OS for running a database.

yum install cloog-ppl compat-libcap1 compat-libstdc++-33 cpp gcc gcc-c++ glibc-devel glibc-headers
 kernel-uek-headers ksh libXmu libXxf86dga libXxf86misc libaio-devel libdmx libstdc++-devel mpfr ppl
 xorg-x11-utils xorg-x11-xauth glibc glibc-common

disable selinux

Just to be sure that selinux doesn't spoil the fun. I will disable it:

[root@gwexample ~]# cat /etc/sysconfig/selinux 
# This file controls the state of SELinux on the system.
# SELINUX= can take one of these three values:
#     enforcing - SELinux security policy is enforced.
#     permissive - SELinux prints warnings instead of enforcing.
#     disabled - No SELinux policy is loaded.
# SELINUXTYPE= can take one of these two values:
#     targeted - Targeted processes are protected,
#     mls - Multi Level Security protection.

Step 1: Make it possible to connect to MS SQL server from Linux

Install the unixODBC Driver manager.

As unixODBC is provided in the public-yum repositories of Oracle we can make our life easy and install the software from the repo.

yum install unixODBC unixODBC-devel

Do not forget the devel package, it contains files like sql.h which are necessary to compile freetds with support for unixODBC!

Installation of freetds

Download freetds

Download the latest version of freetds. I downloaded the freetds-stable.tgz. We will copy this to /tmp/freetds-stable.tgz. Execute the following to extract it and move it to /home/oracle.

#As user oracle
cd /tmp
gunzip -c freetds-stable.tgz | tar -xvf -
mv freetds-0.91/ /home/oracle/

Configure and build freetds

Now we need to configure it:

#As user oracle
cd /home/oracle/freetds-0.91
./configure --prefix=/usr/local/freetds --with-tdsver=8.0 --with-unixodbc=/usr

If you get an error about missing sql.h. It is most likely because unixODBC is installed on a different location. Do a search for the file (either `locate sql.h` or if locate is not available `find / -name "sql.h"`). If the location is /aa/bbb/c/include/sql.h the configure argument should be --with-unixodbc=/aa/bbb/c . So for my setup it was /usr/include/sql.h

Next build the source

#as user oracle

And finally install it

#as root
make install

Configuring unixODBC driver manager

Setup the content of /etc/odbc.ini

[ODBC Data Sources]
#I do not think this line is really necessary but it does no harm :-)
#The name between square brackets should correspond to the DB name of your SQL server.
#From now on it will be referred to as DSN !!
Driver = /usr/local/freetds/lib/
#Do not use Driver = /usr/lib64/
#Location of the MS SQL Server (ip or hostname)
Server =
#Port where MS SQL Server is listening
Port = 1433
TDS_Version = 8.0
TraceFile = /tmp/odbc.trace
Trace = Yes
dump file = /tmp/freetds.log

The following should be checked:

  • You should check if the file /usr/local/freetds/lib/ exists, because without it it won't work!
  • If you use a hostname to locate the server, make sure it is resolvable. Try to do `ping ` and see if you can reach the server. (If not try `ping ip` , if that doesn't work ICMP might be disabled). As I use a simple setup with VMs and no DNS server I use /etc/hosts file of linux to use the hostname.

Verify if freetds can be used via odbc

isql  -v   
| Connected!                            |
|                                       |
| sql-statement                         |
| help [tablename]                      |
| quit                                  |
|                                       |

For example isql -v MSSQLVM test Welcome1

Setup Oracle DB gateway for ODBC

Download software

The Oracle DB gateway for ODBC can be downloaded separately if you have got an Oracle Support Account. Instructions are available at . (you need to get Oracle Gateways)

Install software

Unzip the downloaded zip. And run the installer gateways/runInstaller

In splash screen click next.

We leave the default values for location and name.

No incompatibilities were found -> Next.

Select Oracle Database Gateway for ODBC -> Next

Click the Install button to install the software.

Execute the 1 or 2 scripts mentioned in the dialog screen AS ROOT.

select Perform typical configuration and click Next.

An error might popup. Just click OK

The error has been noted. Click retry to try again.

If we do exactly the same will it work now?

Apparently yes :-).

Verify if the software works

#as user oracle
 export ORACLE_HOME=/home/oracle/product/11.2.0/tg_1

Oracle Corporation --- WEDNESDAY FEB 06 2013 21:02:11.918
Heterogeneous Agent Release - 64bit Production Built with
Oracle Database Gateway for ODBC

Everytime you start dg4odbc make sure ORACLE_HOME is set!

Configure the listener

This must be done on the host ware DG4ODBC is installed. Because this listener will be your entry point to the MS SQL SERVER. Listener config file is situated at $ORACLE_HOME/network/admin/listener.ora
      (SID_DESC =
         (SID_NAME = MSSQLVM)
         (ORACLE_HOME = /home/oracle/product/11.2.0/tg_1)
         (PROGRAM = dg4odbc)
       (ADDRESS_LIST =
          (ADDRESS = (PROTOCOL = TCP)(HOST = gwexample)(PORT = 1511)) 
ADR_BASE_LISTENER = /home/oracle/product/11.2.0/tg_1

Important notes with listener config:

  • ENVS is not a typo, it should be ENVS
  • SID_NAME should correspond to DSN from /etc/odbc.ini
  • Each occurence of '/home/oracle/product/11.2.0/tg_1' should be replaced by the oracle home of dg4odbc but you cannot use the variable $ORACLE_HOME
  • HOST should be set to hostname or IP of the machine that hosts the dg4odbc. (if hostname is used it is best to have it in /etc/hosts of this machine)
  • PORT should be set to any free port
  • Firewall should be configured to allow connections to the PORT selected above -> do this now because you will forget about it!
  • ADR_BASE_LISTENER might not be necessary but I set it to the oracle home of dg4odbc and that does work

Configure tnsnames.ora

This file should be available at each machine that has an Oracle Database that wants to connect to the MS SQL server using the DG4ODBC. It is used to resolve the listener!

tnsnames config file is situated at $ORACLE_HOME/network/admin/tnsnames.ora


Important notes with tnsnames.ora config

  • The name at the start is chosen to be the same as the DSN for convenience
  • The value for HOST does not have to be the ip/hostname of the DB server but must be the ip/hostname of the machine where dg4odbc is installed (this can be the same as the DB server)
  • The value after the SID must be the DSN
  • The line (HS=OK) must be present!

Configure dg4odbc

Configuration of this is done in $ORACLE_HOME/hs/admin/init.ora

# This is a sample agent init file that contains the HS parameters that are
# needed for the Database Gateway for ODBC
# HS init parameters

# ODBC specific environment variables
set ODBCINI=/etc/odbc.ini

# Environment variables required for the non-Oracle system
#set =

Important notes with dg4odbc config

  • HS_FDS_CONNECT_INFO must be set equal to the DSN
  • HS_FDS_TRACE_LEVEL must be set to 0 after setup is working! It should only be set to 255 to do troubleshooting at the beginning.
  • HS_FDS_SHAREABLE_NAME should point to the ODBC library (i.e. the one of the ODBC driver manager)
  • HS_LANGUAGE should be set to american_america.we8iso8859P1

Reload the listener

This should be done for each change of listener.ora (on dg4odbc machine) and tnsnames.ora (on DB machine)

#as user oracle
#Make sure if ORACLE_HOME is set correctly and if DB SID
lsnrctl reload

Test if dg4odbc can be reached from the DB server

#as user oracle
#if not set, set oracle environment using . oraenv
[oracle@dbserver ~]$ tnsping MSSQLVM
TNS Ping Utility for Linux: Version - Production on 05-FEB-2013 10:09:48
Copyright (c) 1997, 2011, Oracle.  All rights reserved.
Used parameter files:

Used TNSNAMES adapter to resolve the alias
Attempting to contact (DESCRIPTION= (ADDRESS=(PROTOCOL=tcp)(HOST=gwexample)(PORT=1511)) (CONNECT_DATA=(SID=MSSQLVM)) (HS=OK))
OK (0 msec)

Creat DB link

#in the Oracle DB execute:
create database link db_link connect to "test" identified by "Welcome1" using 'MSSQLVM';
SELECT * FROM dbo.table@db_link;

Important notes:

  • db_link is the name of the database link. It can be used after an @-sign
  • In the create database link statement BOTH the username and password should be surrounded by double quotes.
  • After using comes the DSN, surrounded by single quotes.


Handy files

  • /tmp/sql.log
  • $ORACLE_HOME/hs/log directory contains traces

Error messages + solution

Strange output -> language issue (we only speak americano)

sqlplus output

ORA-28500: connection from ORACLE to a non-Oracle system returned this message:

ORA-02063: preceding 2 lines from link_name

Sqldeveloper ouput

For Sqldeveloper is the output a bit clearer:

ORA-28500: connection from ORACLE to a non-Oracle system returned this message:
[ u n i x O D B C ) ( D r i v e r M a n a g e r ] D a t a s o u r c e n a m e ...

Strange output comes due to incorrect language settings. Please verify that HS_LANGUAGE=american_america.we8iso8859P1 is set in init.ora !!

Corrupt link

This time no error message is displayed at all.

ORA-28500: connection from ORACLE to a non-Oracle system returned this message:
ORA-02063: preceding 2 lines from link_name
 (and ORA-28500) are caused by invalid 

If the listener is restarted with a new config your link in the database might have become invalid. Drop it and recreate it:

create database link db_link connect to "test" identified by "Welcome1" using 'MSSQLVM';

Nonexistent table or view

ORA-00942: table or view does not exist
[FreeTDS][SQL Server]Invalid object name 'your.table' . {42S02,NativeErr = 208}[FreeTDS][SQL Server]Statement(s) could not be 
ORA-02063: preceding 2 lines from link_name
00942. 00000 - "table or view does not exist"

This can happen if there are quotes surrounding the table name -> Remove surrounding quotes!


Some of the contents is based on

De freetds website . Thank you kgronau for the answer on setting HS_LANGUAGE!