IBM Db2

Overview

This check monitors IBM Db2 through the Datadog Agent.

Setup

Installation

The IBM Db2 check is included in the Datadog Agent package.

Dependencies

The ibm_db client library is required. To install it, ensure you have a working compiler and run:

Unix

sudo -Hu dd-agent /opt/datadog-agent/embedded/bin/pip install ibm_db==3.1.0

Note: If you are on an Agent running Python 2, use ibm_db==3.0.1 instead of ibm_db=3.1.0.

Windows

For Agent versions <= 6.11:

"C:\Program Files\Datadog\Datadog Agent\embedded\python.exe" -m pip install ibm_db==3.0.1

For Agent versions >= 6.12 and < 7.0:

"C:\Program Files\Datadog\Datadog Agent\embedded<PYTHON_MAJOR_VERSION>\python.exe" -m pip install ibm_db==3.0.1

For Agent versions >= 7.0:

"C:\Program Files\Datadog\Datadog Agent\embedded3\python.exe" -m pip install ibm_db==3.1.0

On Linux there may be need for XML functionality. If you encounter errors during the build process, install libxslt-dev (or libxslt-devel for RPM).

Enable monitoring

The IBM Db2 integration pulls data using the following table functions:

• MON_GET_TABLESPACE
• MON_GET_TRANSACTION_LOG
• MON_GET_BUFFERPOOL
• MON_GET_DATABASE
• MON_GET_INSTANCE

For more information about these table functions, see the official IBM documentation.

To monitor a Db2 instance, create a Db2 user with either the EXECUTE permission on the above five table functions, or grant the Db2 user one of the following roles:

• DATAACCESS authority
• DBADM authority
• SQLADM authority

To monitor the health of an instance, its associated databases, and database objects, enable the database system monitor switches for each of the objects you want to monitor:

• Statement
• Lock
• Tables
• Buffer pool

Switch to the instance master user and run these commands at the db2 prompt:

update dbm cfg using HEALTH_MON on
update dbm cfg using DFT_MON_STMT on
update dbm cfg using DFT_MON_LOCK on
update dbm cfg using DFT_MON_TABLE on
update dbm cfg using DFT_MON_BUFPOOL on

Next, run get dbm cfg and you should see the following:

 Default database monitor switches
   Buffer pool                         (DFT_MON_BUFPOOL) = ON
   Lock                                   (DFT_MON_LOCK) = ON
   Sort                                   (DFT_MON_SORT) = OFF
   Statement                              (DFT_MON_STMT) = ON
   Table                                 (DFT_MON_TABLE) = ON
   Timestamp                         (DFT_MON_TIMESTAMP) = ON
   Unit of work                            (DFT_MON_UOW) = OFF
 Monitor health of instance and databases   (HEALTH_MON) = ON

Configuration

Validation

Run the Agent’s status subcommand and look for ibm_db2 under the Checks section.

Data Collected

Metrics

Events

• ibm_db2.tablespace_state_change is triggered whenever the state of a tablespace changes.

Service Checks

ibm_db2.can_connect
Returns CRITICAL if the Agent is unable to connect to the monitored IBM Db2 database, or OK otherwise.
Statuses: ok, critical

ibm_db2.status
Returns CRITICAL if the monitored IBM Db2 database is quiesced, WARNING for quiesce-pending or rollforwards, or OK otherwise.
Statuses: ok, warning, critical, unknown

Troubleshooting

CLI Driver SQL1531N error

If you encounter an issue that produces error logs like the following:

2023-08-10 23:34:47 UTC | CORE | ERROR | (pkg/collector/python/datadog_agent.go:129 in LogMessage) | ibm_db2:c051131490335a94 | (ibm_db2.py:563) | Unable to connect to database `datadog` as user `db2inst1`: [IBM][CLI Driver] SQL1531N  The connection failed because the name specified with the DSN connection string keyword could not be found in either the db2dsdriver.cfg configuration file or the db2cli.ini configuration file.  Data source name specified in the connection string: "DATADOG". SQLCODE=-1531

Then it’s most likely caused by one of the following scenarios:

• The configuration (conf.yaml) is missing a host and port configuration
• The CLI Driver isn’t able to locate the database due to the absence of db2cli.ini and db2dsdriver.cfg

The Agent requires the information in both of the above scenarios to determine where to properly connect to the database. To solve this issue, you can either include a host and port parameter for every instance of the ibm_db2 check experiencing this issue. Alternatively, if you want to use the DSNs defined in either the db2cli.ini or db2dsdriver.cfg files, you can copy those files over to the clidriver directory that the Agent uses. Under normal circumstances, that directory is located at /opt/datadog-agent/embedded/lib/python3.9/site-packages/clidriver/cfg for Linux.

Installing ibm_db client library offline

If you’re in an air gapped environment, or on a restricted network where it’s not possible to run pip install ibm_db==x.y.z where x.y.z is the version number, you can install ibm_db using the following method:

1. On a machine with network access, download the source tarballs for the ibm_db library and the ODBC and CLI. The ODBC and CLI are required to be downloaded separately because the ibm_db library requires them, but it cannot download them via pip. The following script installs the archive file for ibm_db==x.y.z on a Linux machine, where x.y.z is the version number:

   curl -Lo ibm_db.tar.gz https://github.com/ibmdb/python-ibmdb/archive/refs/tags/vx.y.z.tar.gz
   
   curl -Lo linuxx64_odbc_cli.tar.gz https://public.dhe.ibm.com/ibmdl/export/pub/software/data/db2/drivers/odbc_cli/linuxx64_odbc_cli.tar.gz
   

2. Transport the two files over to the restricted host, and then extract the archive.

   tar -xvf ibm_db.tar.gz
   
   tar -xvf linuxx64_odbc_cli.tar.gz
   

3. Set the IBM_DB_HOME environment variable to the location of where /clidriver was extracted from linuxx64_odbc_cli.tar.gz. This will prevent the ibm_db library from installing a new version of the ODBC and CLI since that would fail.

   export IBM_DB_HOME=/path/to/clidriver
   

4. Using the embedded pip on the Agent, install the ibm_db library locally. This library’s files are contained within the extracted python-ibmdb-x.y.z from ibm_db.tar.gz.

   /opt/datadog-agent/embedded/bin/pip install --no-index --no-deps --no-build-isolation  /path/to/python-ibmdb-x.y.z/IBM_DB/ibm_db/
   

If you get the following error:

  error: subprocess-exited-with-error

  × Preparing metadata (pyproject.toml) did not run successfully.
  | exit code: 1
   -> [8 lines of output]
      Detected 64-bit Python
      Detected platform = linux, uname = x86_64
      Downloading https://public.dhe.ibm.com/ibmdl/export/pub/software/data/db2/drivers/odbc_cli/linuxx64_odbc_cli.tar.gz
       Downloading DSDriver from url =  https://public.dhe.ibm.com/ibmdl/export/pub/software/data/db2/drivers/odbc_cli/linuxx64_odbc_cli.tar.gz
      Pre-requisite check [which gcc] : Failed

      No Gcc installation detected.
      Please install gcc and continue with the installation of the ibm_db.
      [end of output]

You may need to install gcc.

Need help? Contact Datadog support.

Further Reading

Additional helpful documentation, links, and articles:

• Monitor IBM DB2 with Datadog