Migrating Pulse to a new server

This article describes the steps to follow when you want to migrate Pulse from an old server to a new server.

1. Backing up Pulse from the old server.

If you want to keep Pulse data from the old server, you will need first to back-up some Pulse folders from the old server: 

2. Install Pulse on the new server.

The installation of Pulse is straight forward, follow the installation wizard.

3. Stop Pulse for TM1 services.

4. Replace the new Pulse folders with the old ones.

Copy the back up folders from step 1 into the same location on the new server.

5. Update Pulse databases if server name or TM1 instance changed.

Do this last step only if the new server name is different from the old one and if the TM1 instance name changed.

The server name and TM1 instance name are stored in the Pulse databases, to update the Pulse databases, steps are explained in this article:

6. Start Pulse for TM1 services

Update the JVM maximum memory

The Java application server (pulseas.exe) which runs in the background has a default memory of 1024 MB.

On a very large TM1 application or on a server with lots of TM1 instances, the JVM of the Pulse application server can reach this value.

Once it has reached this value, Pulse web client will slow down because it has to make some spaces for the new objects by deleting unused objects in order to keep the memory below 1024 MB, this is called garbage collector.

If Pulse web client slows down or if Pulse reports server out of memory errors, you should increase this value. This can be done manually:

  • Open regedit:


  •  Go to HKEY_LOCAL_MACHINE > SOFTWARE > WOW6432Node > Apache Software Foundation > Procrun 2.0 > PulseAppServer > Parameters > Java > JvmMX
  • Double click on JvmMx and tick decimal to see the value:
  • Update the value and click OK.
  • Restart Pulse services



Renaming TM1 server name and instance name

Pulse stores the TM1 server name and instance name in the Pulse databases. Pulse stores the data in the databases by server name and instance name. If you have to rename the server name where TM1 is installed or to rename the TM1 instance name, you will have to update the databases with a script.

You would require to do this if you’re migrating from one box to another where the server name will be different or when upgrading from Cognos Express to TM1 and you don’t want to keep the CXMD instance name.

In order to do that we’ve added 2 new scripts in C:\Program Files\Pulse for TM1\utils which will update the Pulse databases and rename the VCS folder:

  • ServerNameUpdate.bat: rename the server name
  • ServiceNameUpdate.bat: rename the service name

Pulse services have to be stopped before running these batch files.

In order to run a batch file:

  • open the command lines:


  • go to the utils folder cd C:\Program Files\Pulse for TM1\utils:


  • Just type the bat file you want to run e.g ServerNameUpdate.bat to see the help


  • To rename a server name from A to B you'll need to type ServerNameUpdate.bat A B and then press enter (do not forget to stop the Pulse services).

How to create the license optimisation report

Pulse has specially designed report that gives you a summary of how the system is used by different types of users. This report gives you a full picture over any time-frame of how well utilized your licenses are. Users are categorized into Admin, Write and Read-Only roles and their usage mapped according to the amount of time they spend in the system and what they are doing. 
Use this information to determine whether you require more licenses or allow the reallocation of licenses for users who do not use the system.

In order to run the optimization report, you have to open the Pulse Thick Client:

Click on Tools then User Login:

Enter User Name and Password of an admin account:

Go to Tools then License Optimisation:

Select a time frame and then click the refresh button in order to see the report:

Click the Export button to export the report as PDF or Word document:

How Pulse determine which users is Read-Only or NOT:

Each time Pulse updates the documentation it will check first in TM1 if a user has ReadOnlyUser value equal to 1 (}ClientProperties cube) then User is set to Read-Only else Pulse will loop through user/group/cube security, if there is anything that isn’t read its sets Read-Only to False.

What's inside the dump file

A dump file is a zip file which contains history of data of the previous 10 min of your TM1 instance or TM1 server. You'll find details of everything that happened in the previous 10 min before the dump file was created such as who was logged in, which processes or chores ran during the past 10 min, snapshots of TM1 threads...

The dump file is very helpful for example if the TM1 instance crashes and you want to know who was logged in or what was running. The dump file will help you to analyze in details the cause of the server crash.

Pulse creates a dump file every time a TM1 instance stops or if an alert is triggered with the option "Create Dump File" ticked:

All dump files are created in the dump folder: C:\Program Files\Pulse for TM1\dump.

They are just zip files, you can unzip them with 7-zip or other zip programs, just right click and unzip it:

It contains several files with details of the previous 10 min of history of your TM1 server:

  • cxmd (<instance name>): tm1s.cfg + TM1 error logs.
  • chores.csv: The list of chores which were running in the past 10 min.
  • dump.json: General information on the dump file (when it was created..).
  • history.json: The history of all instances' state in a JSON format.
  • logs.csv: The last 10 min of TM1 message logs.
  • operations.csv: List of operations (TM1 process, cube view...) which took more than 10 sec during the past 10 min .
  • processes.csv: List of processes which were running in the past 10 min.
  • server.csv: Server stats (Memory and CPU usage) from the past 10 min.
  • service.csv: Instance stats from the past 10 min.
  • servicestate.json: Service states from the past 10 min.
  • sessions.csv: Session stats from the past 10 min.
  • state.json: The last instances states (Loading, Running...).
  • top.csv: A snapshot of TM1 threads every sec for the last 10 min.