Your help desk application may have the facility to store a large amount of data but this can also hinder the performance of your help desk tool. This guide provides certain queries to improve the performance of ManageEngine ServiceDesk Plus.
To execute the query, you need to access the MySQL database. Click the respective OS link (Windows or Linux) to know how to access the MySQL database.
NOTE: Please restart ServiceDesk Plus after executing the query for the changes to get applied.
The list of performance tips discussed in this guide is,
It is required to have a 64bit server machine. The default configuration set in wrapper.conf is sufficient. However, in a larger environment, set the max heap size as follows,
# Maximum Java Heap Size (in MB)
wrapper.java.maxmemory=4096
Enabling MS SQL Snapshot Mode
We can enable the Read committed Snapshot in the MSSQL environment to increase the performance of the application. It creates a snapshot or read-only database of your current results that is separate from your live database. So when you run a SELECT statement, to read your data, you are reading from a read-only copy of your database. When you change your database, it happens on the live database, and then a new copy or snapshot is created for reading against.
All MSSQL databases usually have lock issues, which may affect an application’s optimal performance. To avoid such performance issues in ServiceDesk Plus, enable the read committed snapshot isolation (RCSI) by following these steps:
Check if you have enabled RCSI successfully by running the following query.
You can now start the application. For more information on MSSQL databases and RSCI, go here.
Benefits of enabling the MS SQL Snapshot Mode
Simultaneous user querying of database will be possible without any delay or disruption (although load on application might exceed the usual limits).
User querying records will be maintained in a separate tempdb, which will not be affected by other user queries occurring in parallel or subsequently.
Database tables will be uploaded only after a transaction reaches completion.
If you have many nodes say, more than 1000, then scanning these nodes at regular intervals would hinder the performance of ServiceDesk Plus. To reduce the load on the server, you could scan these nodes using Distributed Asset Scan. All you need to do is install remote ServiceDesk Plus servers across departments. These remote servers scan the nodes at regular intervals and export the data to the central ServiceDesk Plus server.
The Request Count displays the total number of requests in the request list view page. Larger the request count value, the longer it takes to display the requests in the list view page.
The request count value cannot be removed. Instead you can choose to view the request count, by clicking a Row Count button in the request list view page.
To show the Row Count button, use the query,
The Refresh Timer refreshes the request list view page at regular intervals. But this can slow down the performance of ServiceDesk Plus.
You can disable the Refresh Timer option by using the query,
The 'All Requests' filter in the request list view displays all the requests that have been created so far, irrespective of the status. But as the number of requests increases, the performance of ServiceDesk Plus gradually slows down.
In this case, you can remove the 'All Requests' option from the filter drop-down using the query,
In the Request List View page, you would find a drop-down menu to select the number of requests to view per page. Limiting this number to 25 or 50 would eliminate loading of requests and improve the performance of ServiceDesk Plus.
Short Description refers to the description which appears when you hover over the subject link of the request in the request list view. By default, while performing a search operation, the short description of a request would be searched. But when you have large amount of data, the performance of ServiceDesk Plus eventually decreases.
You can disable this feature by using the query,
By default, the recent items list is deleted once in every 15 days. But you can improve ServiceDesk Plus's performance by increasing the cleanup frequency.
Example: If you want to cleanup the Recent Item list every 5 days, then use the query,
The maximum limit for cleaning up recent item list is 90. If you want to disable the cleanup, then set the parameter value as -1.
By default, the error log list is deleted once in every 180 days. But you can increase the cleanup frequency to accelerate the backup process.
Example: If you want to cleanup the Error Log list every 30 days, then use the query,
The maximum limit for cleaning up error log list is 365. If you want to disable the cleanup, then set the parameter value to -1.
It is a table containing the session details i.e., the logged in and logged out information. These entries are not used by the application and can be removed periodically for better database performance. By default, the session details are deleted once in every 90 days but for better performance, you can increase the cleanup frequency.
Example: If you want to cleanup the ACC Session every 30 days then use the query,
The maximum limit for cleaning up the ACC session details is 365. If you want to disable the cleanup, then set the parameter as -1.
System Generated Notifications are notifications generated and sent by the system. You can either delete all the system notifications or manually check the notifications to delete.
If you want to delete all the system notifications then execute the query,
delete from notification where senderid=1;
To get the list of notification titles to delete unwanted notifications, execute the query,
select notificationtitle from notification limit 100;
Example: If the notificationtitle is 'has been added to the group', then use the query to delete notifications under this title,
delete from notification where notificationtitle like '%has been added to the group%';
By default, the number of user data objects that would be cached is 500. But in case of high-end machines with better RAM configurations, this value can be increased to cache more data for quick response.
Example: If you want to increase the cache count to 1000, then use the query below.
For build 11000 or newer:
For older builds:
By default, the number of technician data objects that would be cached is 300. But in case of high-end machines with better RAM configurations, this value can be increased to cache more data for quick response.
Example: If you want to increase the cache to 1000, then use the query below.
For build 11000 or newer:
For older builds:
By default, the number of Message IDs that would be cached is 1000. But in case of high-end machines with better RAM configurations, this value can be increased to cache more data for quick response.
Example: If you want to increase the cache count to 2000, then use the query,
By default, the number of E-mail IDs / User IDs that would be cached is 1000. But in case of high-end machines with better RAM configurations, this value can be increased to cache more data for quick response.
Example: If you want to increase the cache count to 2000, then use the query,
Log File Size Threshold Violation (MSSQL): Best Practices
The recovery model of a database defines how transactions are logged and retained.
There are 3 recovery models in use: Simple, Full, and Bulk Logged. Here, we will discuss only Simple and Full models.
In SQL Server, transaction log files, when not backed up regularly, can grow and cause a database failure because the recovery model, by default, is Full.
To prevent this scenario, DBAdmins can perform:
a) Full recovery model with regular transaction log backups
or
b) Set the database recovery model to Simple.
In the Simple recovery model, point in time recovery is not possible, whereas it is possible in the Full recovery model.
Therefore, in a production environment, configure your database with the Full recovery model, so that a point in time recovery option is easily available.
Follow these steps to create a backup log file:
Output:
Output
In the SQL Server Management Studio
i) Under Database files and filegroups choose File type as Log.
ii) Under Shrink action choose Release unused space
Each time the application server experiences performance issues, the CPU and memory usage statistics will be closely monitored and documented. This is applicable only for Windows.
Enable the monitoring script to record the CPU and memory usage statistics. Execute the following commands in PowerShell.
Go to ServiceDesk Home/tools/monitoring/CPU_Memory_Monitoring.ps1 and unblock the PowerShell script located in the application directory.
Set the execution policy to run the custom script. You can use one of the following execution policies: Bypass, Unrestricted, or RemoteSigned.