Troubleshooting problems in AWS Parallel Computing Service

The following topics provide guidance to troubleshoot some problems you might encounter in AWS PCS.

• Cluster updates

• Cluster version updates

• Compute node bootstrap problems

• Custom Slurm settings

• EC2 instances terminated after reboot

• Identity and access

• Invalid node registration (INVALID_REG)

• Job submission MaxJobCount limit

• Missing accounting records

• Slurm reboot issues