Best practices for preventing cache memory issues

As discussed below, you can prevent the following AR System server processes from reaching its maximum size in memory, which can cause the process to terminate or to stop responding:

• arserverd (UNIX)
• arserver.exe (Windows)

For more information about resolving cache memory issues, see Memory leak or Admin Changes on the BMC Community.

Configuring the AR System server to control memory use

To maximize memory use in your environment, implement the following AR System server configuration best practices:

• Limit large queries.
• Do not allow users to perform unqualified searches. To remove all unqualified searches, review existing workflow and modify it as necessary.

• Set the following AR System server configuration options appropriately:

  Option	Description
  Max-Entries-Per-Query	For details about this setting, see Max-Entries-Per-Query.
  Cache-Display-Properties	For details about this setting, see Cache-Display-Properties.
  Disable-ARSignals	For details about this setting, see Disable-ARSignals.

Avoid making administrative changes during peak hours

Do not make administrative changes during periods of peak usage. 

This ban includes actions in BMC Helix ITSM applications that can cause an admin copy cache or a client cache load (see Actions-in-BMC-Helix-ITSM-applications-that-trigger-caching). 

Identify a time period when administrative changes can be performed, and ensure that all developers and application administrators understand the importance of making administration changes during this period only.

Checking the AR System log files for long-running operations

Some long-running operations (such as escalations, archiving, or API calls) prevent the copy of the cache that they use from being freed until the operation is finished. Multiple long-running operations can tie up two or more copies of the cache. Check the AR System log files for such activity, and avoid it during peak hours. For more information, see Enabling-and-analyzing-logs.