Limits
- 1 Summary
- 2 Cache Limits
- 2.1 Cache Sizes - Limit
- 2.2 Cache Retention - Expiry
- 2.3 Applying change
- 2.4 Modifiable Caches
- 2.4.1 featureConfCache
- 2.4.2 ldapDistGroupMembersCache
- 2.4.3 notifHistEnabledCache
- 2.4.4 profileUpdatedInfoCache
- 2.4.5 templateCache
- 2.4.6 usageDailyCache
- 2.4.7 usageHourlyCache
- 2.4.8 usageMonthlyCache
- 2.4.9 userMacroCache
- 2.5 Resetting caches back to Default Values
- 2.6 Error Banner - The JEMH Limits cache section has changed. Cluster restart is required to apply.
Summary
since 4.1.31
This feature enables the customization of various configuration parameters ‘Limits' within JEMH that can allow admins to optimise settings for performance (more caches = more memory but lower database usage) vs resource use (less caches = less memory but more database usage).
In this release we provide the basic feature and control over all DC clusters caches within JEMH. Future releases will expand this to cover other areas and other parameters (eg to limit the number of Profiles, Project Mappings & Rules that can be created).
Cache Limits
Cache Sizes - Limit
Caches in the limits section represent data shared across the DC cluster, higher sizes will consume more memory (where data is sufficient to drive such usage) but will reduce database access. Values set too low can result in a slower user interface as well as generally increased database activity.
Cache Retention - Expiry
Retention windows allow cached data to expire. Data is re-read on demand and re-added to the cache, so, a balance can be set between retaining data that ‘might’ be needed again vs memory usage. Re-calculating and caching data can be expensive (eg distribution list membership), but, that distribution group list may change, so having a retention period too high may delay JEMH being able to reflect that change.
The default retention is 1h, a higher value, eg 1day may/not be a valid choice in your environment, this is why the feature exists.
Applying change
Cache changes will require a Cluster rolling restart to apply.
Modifiable Caches
Press the Edit button to modify the Cache Limits. This will then show the below page.
featureConfCache
Cache of enabled feature configuration (1 only). E.g. Ad-Hoc Notifications.
Default Limit Value - 1
Default Expiry (hours) Value - 1
ldapDistGroupMembersCache
Contains the members of email distribution groups extracted via LDAP.
Default Limit Value - 1000
Default Expiry (hours) Value - 1
notifHistEnabledCache
Cache of notification history enablement (1 only). Notification History is enabled within Notifications > Notification Mapping > Store Notification History.
Default Limit Value - 1
Default Expiry (hours) Value - 1
profileUpdatedInfoCache
Contains the timestamp information relating to Profiles (Profiles are cached separately per node).
Default Limit Value - 50
Default Expiry (hours) Value - 1
templateCache
Contains the Template Sets that are used for notifications.
Default Limit Value - 1000
Default Expiry (hours) Value - 1
usageDailyCache
Contains the daily usage
Default Limit Value - 1000
Default Expiry (hours) Value - 1
usageHourlyCache
Contains the Hourly usage
Default Limit Value - 1000
Default Expiry (hours) Value - 1
usageMonthlyCache
Contains the Monthly usage
Default Limit Value - 1000
Default Expiry (hours) Value - 1
userMacroCache
Contains user defined velocity macro script (Script found within Template Sets > Macros > User Macros).
Default Limit Value - 1000
Default Expiry (hours) Value - 1
Resetting caches back to Default Values
To Reset the Cache Limits back to Default, you will have to Press Edit and then select Apply all Defaults. Note: This would still require a Rolling Cluster Restart to apply.
Error Banner - The JEMH Limits cache section has changed. Cluster restart is required to apply.
When the Cache limits are modified it will result in the above banner being shown to highlight that not all of the Cluster nodes have been restarted yet and are currently not using the new Cache limits. To resolve this you would need to restart each node so that they have the new Cache Limits.
You can view the cluster nodes that require a restart by going to JEMH > Limits and then viewing the Clustering section as this will show all of the configured nodes and will identify which nodes require a Reboot: e.g. below shows that node2 needs a reboot.