This section describes logged issues that you might encounter when running deployments.
The Master Server has run out of memory during a deployment. You might find this out-of-memory error in the master server's log file, or your might find a similar error message in the browser interface. This error can be caused by deploying large, complex components or deploying to several hosts at the same time.
Examples of large and complex deployments:
Deploying a container that has more than 100 referenced components.
Deploying a system#directory component that has a number of large files. Files that are bigger than 250k are considered large files.
Deploying many plans to many hosts.
An example of this is deploying four or more plans to four or more host sets where each host set contains eight or more hosts and each plan deploys more than 100 resources.
You can prevent this error in two ways. First, you can perform less complex deployments with fewer hosts and components. Second, you can increase the physical RAM in the Master Server and then increase the CR_MAX_HEAP_SIZE value.
Ensure that your system has enough RAM before increasing the CR_MAX_HEAP_SIZE value.
To establish the new CR_MAX_HEAP_SIZE value, determine the amount of unused physical RAM that was available when the system ran out of memory. The maximum value for CR_MAX_HEAP_SIZE should be 100Mb less than the amount of unused physical RAM.
To set the value, modify the /N1SPS-home/server/bin/cr_server script by changing the CR_MAX_HEAP_SIZE value to the following.