Introduction
Are you encountering any unexpected situations while working on Intel® Developer Cloud for oneAPI? This article describes the self-help collection of troubleshooting techniques for Intel Developer Cloud for oneAPI, which describes a problem and countermeasures for several such situations.
Connectivity Issues
Connectivity issues could be due to multiple factors. The following are a few expected reasons for losing access to Intel Developer Cloud for oneAPI:
- Account deactivated due to prohibited activity.
- Account expired.
Deactivated Due to Any Prohibited Activity
Error: Your DevCloud account has been deactivated due to observed prohibited use such as cryptocurrency mining, data scraping, malicious program or tampering with DevCloud services.
In our efforts to thwart the progress of cryptocurrency miners and keep Intel Developer Cloud compute resources free and available to our valued developers, we have had to take both automated and manual action against unauthorized activities by terminating or deactivating accounts.
Unfortunately, we are aware that, while infrequent, several valid accounts have been terminated inadvertently during this process. We apologize for any inconvenience it may have caused. We are continually working to avoid this and create new capabilities for account recovery.
Intel Developer Account Expiry
To check your Intel Developer Cloud account expiry date:
- Log in to https://devcloud.intel.com/oneapi/documentation/shell-commands/
- After you log in, your account expiry date is shown on the top right corner.
To extend your Intel Developer Cloud account access:
If your account access expiry date is nearing, reach out to us one week in advance for an access extension request.
You can also request access extension through the Request Extension link.
If this method does not work, please submit a forum ticket.
If Your Intel Developer Cloud Account Has Expired
Ensure that you have backed up all the data in Intel Developer Cloud. If your account is expired, we won’t be able to extend access or retrieve it back and suggest you have regular backups. Intel and the Intel Developer Cloud team are not responsible for any loss of data. You have to enroll here for a new Intel Developer Cloud account.
Jupyter* Notebook Related Issues
Jupyter* Notebook – Internal Server Error
This might be due to Intel Developer Cloud being very busy or no Jupyter Hub slots being available. Try again after some time.
Jupyter Notebook – Time-out Error
Note that the maximum wall time is 24 hours for .ssh and four hours for Jupyter Notebook. Jupyter Notebook sessions cannot be increased beyond four hours to ensure fair utilization of resources by all. Instead, you could put jobs in the queue and set the wall time to the maximum 24 hours by running the following command:
qsub […] -l walltime=hh:mm:ss <job script>
Issues Related to Jobs on Intel Developer Cloud
Unable to Run Nodes on Log In Node Error: Resource Limit Exceeded
Error:
Process exceeded login node resource limit.
Please run jobs on compute nodes using qsub or start an interactive shell using qsub -I.
Please visit the forums at: https://software.intel.com/en-us/forums/intel-devcloud
The log in node is lightweight and not capable of handling heavy workloads. It is primarily intended to save your data. Hence, a limit is enforced on both memory and computer on log in nodes. This ensures that a memory error is shown if you try to run any heavy tasks on the login node. Try logging in to a computer node to run heavy workloads with the following commands:
qsub -I
qsub -l nodes=<property name>:ppn=2 -d . <job script>
Premature Job Termination
To ensure fair utilization of Intel Developer Cloud resources for all users, there is a maximum limit on the number of jobs that can be run by a specific user. This may be the reason for your issue. The limit of total running jobs per user is two, and each job can request up to two nodes.
Job Stuck in a Queue
If a job is stuck in a queue when there are no nodes free to run the job, try submitting the job again after some time.
Root Access Denied Error
Error: This server does not allow you to run "sudo".
One could observe this error as users do not have root access in Intel Developer Cloud. Intel Developer Cloud is a shared environment, which comes with preinstalled validated Intel® oneAPI software and complementary packages. As a policy, we do not install custom (open source or third-party licensed) software in the environment.
For more details visit: https://devcloud.intel.com/oneapi/get_started/