Troubleshooting Compute Instances on Snowcone Devices
Following, you can find troubleshooting tips for Snowcone jobs with compute instances.
Topics
Virtual Network Interface Has an IP Address of 0.0.0.0
This issue can occur if the physical network interface (NIC) you associated with your virtual network interface (VNIC) also has an IP address of 0.0.0.0. This effect can happen if the NIC wasn't configured with an IP address (for instance, if you've just powered on the device). It can also happen if you're using the wrong RJ45 interface. The Snowcone has two RJ45 interfaces, you may be specifying the wrong physical interface
Action to Take
If this occurs, you can do the following:
-
Create a new VNIC, associated with a NIC that has an IP address. For more information, see Network Configuration for Compute Instances.
-
Update an existing VNIC. For more information, see Updating a Virtual Network Interface.
Snowcone Hangs When Launching a Large Compute Instance
It can appear that your Snowcone has stopped launching an instance. This is generally
not the case. However, it can take an hour or more for the largest compute instances
to launch. You can check the status of your instances using the Amazon CLI command
aws ec2 describe-instances
run against the HTTP or HTTPS Amazon EC2
endpoint on the Snowcone.
My Instance Has One Root Volume
Instances have one root volume by design. All sbe
instances have a
single root volume.
For additional information around adding additional volumes, see https://docs.aws.amazon.com/snowball/latest/snowcone-guide/snowcone-snowcone-ebs.html
Unprotected Private Key File Error
This error can occur if your .pem file on your compute instance has insufficient read/write permissions.
Action to Take
You can resolve this by changing the permissions for the file with the following procedure:
-
Open a terminal and navigate to the location that you saved your .pem file to.
-
Enter the following command.
chmod 400
filename.pem