Beginning in CA Service Desk Manager (CA SDM) 12.9 a popup message to inform end users when the completion of a process is taking longer than expected was introduced. By default a "Delayed Server Response" message will display after 20 seconds to inform the user of a delay. It is obviously in the best interest to remove all other possible problems and roadblocks from the environment to help reduce the number of times this message is seen by end users, but it may come to a point where the architecture of the environment will just not allow any additional refinement and delays will be expected. This document provides a solution to extend the time frame for the popup to appear or to disable it completely so end users will see it less often in environments where delays are to be expected.

Release: 12.6 or higher
Component:  CA Service Desk Manager

This message occurs when it takes longer than the expected 20 seconds for Service Desk Manager to complete a task. An example of delay might be a database query taking longer than expected to return results.

The value "NX_EXPECTED_WEB_RESPONSE" can be introduced into your NX.env file to either increase the timeout or disable it completely. You must add the variable to the NX.env template file so that when pdm_configure is run, the value in the NX.env will not be reset or overwritten.

To increase the threshold to 1 minute follow these steps:

To add the variable to the NX.env run the following command: 

pdm_options_mgr -c -s EXPECTED_WEB_RESPONSE -v 60 -a pdm_option.inst 

To add the variable to the NX.env_nt.tpl file run the following command: 

pdm_options_mgr -c -s EXPECTED_WEB_RESPONSE -v 60 -a pdm_option.inst –t

To disable the threshold follow these steps:

To add the variable to the NX.env run the following command: 

pdm_options_mgr -c -s EXPECTED_WEB_RESPONSE -v 0 -a pdm_option.inst 

To add the variable to the NX.env_nt.tpl file run the following command:

pdm_options_mgr -c -s EXPECTED_WEB_RESPONSE -v 0 -a pdm_option.inst –t

Considerations for Conventional and Advanced Availability Architecture:

Whether you are using a conventional architecture, with a single machine, or primary/secondary configuration, or you are using an advanced availability architecture, with a background, standby, and multiple application servers, you need to run both commands (to add the variable to the NX.env and NX.env template files) on all machines in the environment where CA Service Desk Manager is running. After running the commands on all servers, you must recycle the CA Service Desk Manager services as a whole to invoke the change.

The addition of this NX.env variable and associated value should never be used to mask a real performance problem in the environment. Please make sure that a proper review of the environment and architecture is performed to ensure that everything else is functioning as expected and that there is nothing else that can be done to improve the performance of the CA SDM environment (especially in regards to network connection speeds). This option should only be used if there is a true environmental limitation that cannot be overcome.

Some items to consider when reviewing the performance of an environment:

• Running regular archive/purge rules on items such as tickets, notification log etc.
• Avoid allowing users have the ability to search on fields such as description, activity logs, or any other field where there is a large amount of text to parse
• Perform standard maintenance on the database servers (usually completed by a DBA at regularly scheduled intervals

How may we identify performance issues in Service Desk and what type of data is helpful to Support to resolve these issues?