Using pcluster update - AWS ParallelCluster

Using pcluster update

Starting with AWS ParallelCluster version 2.8.0, pcluster update analyzes the settings used to create the current cluster and the settings in the configuration file for issues. If any issues are discovered, they are reported, and the steps to take to fix the issues are displayed. For example, if the compute_instance_type setting is changed to a different instance type, the compute fleet must be stopped before an update can proceed. This issue is reported when it is discovered. If no blocking issues are reported, you are prompted whether you want to apply the changes.

The documentation for each setting defines the update policy for that setting.

Update policy: These settings can be changed during an update.
Update policy: This setting can be changed during an update.

These settings can be changed, and the cluster can be updated using pcluster update.

Update policy: If this setting is changed, the update is not allowed.

These settings can't be changed if the existing cluster hasn't been deleted. Either the change must be reverted or the cluster must be deleted (using pcluster delete), and then a new cluster created (using pcluster create) in the old cluster's place.

Update policy: This setting is not analyzed during an update.

These settings can be changed, and the cluster updated using pcluster update.

Update policy: The compute fleet must be stopped for this setting to be changed for an update.

These settings cannot be changed while the compute fleet exists. Either the change must be reverted or the compute fleet must be stopped (using pcluster stop), updated (using pcluster update), and then a new compute fleet created (using pcluster start).

Update policy: This setting can't be decreased during an update.

These settings can be changed, but they cannot be decreased. If these settings must be decreased, it is necessary to delete the cluster (using pcluster delete), and create a new cluster (using pcluster create).

Update policy: Reducing the size of a queue below the current number of nodes requires that the compute fleet be stopped first.

These settings can be changed, but if the change would reduce the size of the queue below the current size, the compute fleet must be stopped (using pcluster stop), updated (using pcluster update), and then a new compute fleet created (using pcluster start).

Update policy: Reducing the number of static nodes in a queue requires that the compute fleet be stopped first.

These settings can be changed, but if the change would reduce the number of static nodes in the queue below the current size, the compute fleet must be stopped (using pcluster stop), updated (using pcluster update), and then a new compute fleet created (using pcluster start).

Update policy: If this setting is changed, the update is not allowed. Updating this setting cannot be forced.

These settings can't be changed if the existing cluster hasn't been deleted. Either the change must be reverted or the cluster must be deleted (using pcluster delete), and then a new cluster created (using pcluster create) in the old cluster's place.

This example demonstrates a pcluster update with some changes that block the update.

$ pcluster update Validating configuration file /home/username/.parallelcluster/config... Retrieving configuration from CloudFormation for cluster test-1... Found Changes: # section/parameter old value new value -- ------------------------ ------------------------ ------------------------ [cluster default] 01* compute_instance_type t2.micro c4.xlarge 02* ebs_settings ebs2 - [vpc default] 03 additional_sg sg-0cd61884c4ad16341 sg-0cd61884c4ad11234 [ebs ebs2] 04* shared_dir shared my/very/very/long/sha... Validating configuration update... The requested update cannot be performed. Line numbers with an asterisk indicate updates requiring additional actions. Please look at the details below: #01 Compute fleet must be empty to update "compute_instance_type" How to fix: Make sure that there are no jobs running, then run the following command: pcluster stop -c $CONFIG_FILE $CLUSTER_NAME #02 Cannot add/remove EBS Sections How to fix: Revert "ebs_settings" value to "ebs2" #04 Cannot change the mount dir of an existing EBS volume How to fix: Revert "my/very/very/long/shared/dir" to "shared" In case you want to override these checks and proceed with the update please use the --force flag. Note that the cluster could end up in an unrecoverable state. Update aborted.