Troubleshooting Amazon IoT Greengrass V2

Use the troubleshooting information and solutions in this section to help resolve issues with Amazon IoT Greengrass Version 2.

View Amazon IoT Greengrass Core software and component logs

The Amazon IoT Greengrass Core software writes logs to the local file system that you can use to view real-time information about the core device. You can also configure core devices to write logs to CloudWatch Logs, so you can remotely troubleshoot core devices. These logs can help you identify issues with components, deployments, and core devices. For more information, see Monitor Amazon IoT Greengrass logs.

Amazon IoT Greengrass Core software issues

Troubleshoot Amazon IoT Greengrass Core software issues.

Unable to set up core device

If the Amazon IoT Greengrass Core software installer fails and you aren't able to set up a core device, you might need to uninstall the software and try again. For more information, see Uninstall the Amazon IoT Greengrass Core software.

Unable to start the Amazon IoT Greengrass Core software as a system service

If the Amazon IoT Greengrass Core software fails to start, check the system service logs to identify the issue. One common issue is where Java isn't available on the PATH environment variable (Linux) or PATH system variable (Windows).

Unable to set up nucleus as a system service

You might see this error when the Amazon IoT Greengrass Core software installer fails to set up Amazon IoT Greengrass as a system service. On Linux devices, this error typically occurs if the core device doesn't have the systemd init system. The installer can successfully set up the Amazon IoT Greengrass Core software even if it fails to set up the system service.

Do one of the following:

Unable to connect to Amazon IoT Core

You might see this error when the Amazon IoT Greengrass Core software can't connect to Amazon IoT Core to retrieve deployment jobs, for example. Do the following:

Out of memory error

This error typically occurs if your device doesn't have sufficient memory to allocate an object in the Java heap. On devices with limited memory, you might need to specify a maximum heap size to control memory allocation. For more information, see Control memory allocation with JVM options.

Unable to install Greengrass CLI

You might see the following console message when you use the --deploy-dev-tools argument in your installation command for Amazon IoT Greengrass Core.

Thing group exists, it could have existing deployment and devices, hence NOT creating deployment for Greengrass first party dev tools, please manually create a deployment if you wish to

This occurs when the Greengrass CLI component is not installed because your core device is a member of a thing group that has an existing deployment. If you see this message, you can manually deploy the Greengrass CLI component (aws.greengrass.Cli) to the device to install the Greengrass CLI. For more information, see Install the Greengrass CLI.

User root is not allowed to execute

You might see this error when the user that runs the Amazon IoT Greengrass Core software (typically root) doesn't have permission to run sudo with any user and any group. For the default ggc_user system user, this error looks like the following:

Sorry, user root is not allowed to execute <command> as ggc_user:ggc_group.

Check that your /etc/sudoers file gives the user permission to run sudo as other groups. The permission for the user in /etc/sudoers should look like the following example.

root    ALL=(ALL:ALL) ALL

com.aws.greengrass.lifecyclemanager.GenericExternalService: Could not determine user/group to run with

You might see this error when the core device tries to run a component, and the Greengrass nucleus doesn't specify a default system user to use to run components.

To fix this issue, configure the Greengrass nucleus to specify the default system user that runs components. For more information, see Configure the user that runs components and Configure the default component user.

Failed to map segment from shared object: operation not permitted

You might see this error when the Amazon IoT Greengrass Core software fails to start because the /tmp folder is mounted with noexec permissions. The Amazon Common Runtime (CRT) library uses the /tmp folder by default.

Do one of the following:

Failed to set up Windows service

You might see this error if you install the Amazon IoT Greengrass Core software on a Microsoft Windows 2016 device. The Amazon IoT Greengrass Core software is not supported on Windows 2016, for a list of supported operating systems, see Supported platforms.

If you must use Windows 2016, you can do the following:

com.aws.greengrass.util.exceptions.TLSAuthException: Failed to get trust manager

You might see this error when you install the Amazon IoT Greengrass Core software without a root certificate authority (CA) file.

2022-06-05T10:00:39.556Z [INFO] (main) com.aws.greengrass.lifecyclemanager.Kernel: service-loaded. {serviceName=DeploymentService}
2022-06-05T10:00:39.943Z [WARN] (main) com.aws.greengrass.componentmanager.ClientConfigurationUtils: configure-greengrass-mutual-auth. Error during configure greengrass client mutual auth. {}
com.aws.greengrass.util.exceptions.TLSAuthException: Failed to get trust manager

Check that you specify a valid root CA file with the rootCaPath parameter in the configuration file that you provide to the installer. For more information, see Install the Amazon IoT Greengrass Core software.

com.aws.greengrass.deployment.IotJobsHelper: No connection available during subscribing to Iot Jobs descriptions topic. Will retry in sometime

You might see this warning message when the core device can't connect to Amazon IoT Core to subscribe to deployment job notifications. Do the following:

software.amazon.awssdk.services.iam.model.IamException: The security token included in the request is invalid

You might see this error when you install the Amazon IoT Greengrass Core software with automatic provisioning, and the installer uses an Amazon session token that isn't valid. Do the following:

You can also install the Amazon IoT Greengrass Core software without providing Amazon credentials. For more information, see Install Amazon IoT Greengrass Core software with manual resource provisioning or Install Amazon IoT Greengrass Core software with Amazon IoT fleet provisioning.

software.amazon.awssdk.services.iot.model.IotException: User: <user> is not authorized to perform: iot:GetPolicy

You might see this error when you install the Amazon IoT Greengrass Core software with automatic provisioning, and the installer uses Amazon credentials that don't have the required permissions. For more information about the permissions that are required, see Minimal IAM policy for installer to provision resources.

Check the permissions for the credentials' IAM identity, and grant the IAM identity any required permissions that are missing.

Error: com.aws.greengrass.shadowmanager.sync.model.FullShadowSyncRequest: Could not execute cloud shadow get request

You might see this error when you use the shadow manager component to sync device shadows with Amazon IoT Core. The HTTP 403 status code indicates that this error occurred because the core device's Amazon IoT policy doesn't grant permission to call GetThingShadow.

com.aws.greengrass.shadowmanager.sync.model.FullShadowSyncRequest: Could not execute cloud shadow get request. {thing name=MyGreengrassCore, shadow name=MyShadow}
2021-07-14T21:09:02.456Z [ERROR] (pool-2-thread-109) com.aws.greengrass.shadowmanager.sync.SyncHandler: sync. Skipping sync request. {thing name=MyGreengrassCore, shadow name=MyShadow}
com.aws.greengrass.shadowmanager.exception.SkipSyncRequestException: software.amazon.awssdk.services.iotdataplane.model.IotDataPlaneException: null (Service: IotDataPlane, Status Code: 403, Request ID: f6e713ba-1b01-414c-7b78-5beb3f3ad8f6, Extended Request ID: null)

To sync local shadows with Amazon IoT Core, the core device's Amazon IoT policy must grant the following permissions:

Check the core device's Amazon IoT policy, and add any required permissions that are missing. For more information, see the following:

Operation aws.greengrass#<operation> is not supported by Greengrass

You might see this error when you use an interprocess communication (IPC) operation in a custom Greengrass component, and the required Amazon-provided component isn't installed on the core device.

To fix this issue, add the required component as a dependency in your component recipe, so the Amazon IoT Greengrass Core software installed the required component when you deploy your component.

java.io.FileNotFoundException: <stream-manager-store-root-dir>/stream_manager_metadata_store (Permission denied)

You might see this error in the stream manager log file (aws.greengrass.StreamManager.log) when you configure stream manager to use a root folder that doesn't exist or have the correct permissions. For more information about how to configure this folder, see stream manager configuration.

com.aws.greengrass.security.provider.pkcs11.PKCS11CryptoKeyService: Private key or certificate with label <label> does not exist

This error occurs when the PKCS#11 provider component can't find or load the private key or certificate that you specify when you configure the Amazon IoT Greengrass Core software to use a hardware security module (HSM). Do the following:

Check the documentation for your HSM to learn how to query details about the security tokens in the HSM. If you need to change the slot, object label, or object ID for a security token, check the documentation for your HSM to learn how to do so.

software.amazon.awssdk.services.secretsmanager.model.SecretsManagerException: User: <user> is not authorized to perform: secretsmanager:GetSecretValue on resource: <arn>

This error can occur when you use the secret manager component to deploy an Amazon Secrets Manager secret. If the core device's token exchange IAM role doesn't grant permission to get the secret, the deployment fails and the Greengrass logs include this error.

software.amazon.awssdk.services.secretsmanager.model.SecretsManagerException: Access to KMS is not allowed

This error can occur when you use the secret manager component to deploy an Amazon Secrets Manager secret that is encrypted by an Amazon Key Management Service key. If the core device's token exchange IAM role doesn't grant permission to decrypt the secret, the deployment fails and the Greengrass logs include this error.

To fix the issue, add the kms:Decrypt permission to the core device's token exchange role. For more information, see the following:

java.lang.NoClassDefFoundError: com/aws/greengrass/security/CryptoKeySpi

You might see this error when you try to install the Amazon IoT Greengrass Core software with hardware security and you use an earlier Greengrass nucleus version that doesn't support hardware security integration. To use hardware security integration, you must use Greengrass nucleus v2.5.3 or later.

com.aws.greengrass.security.provider.pkcs11.PKCS11CryptoKeyService: CKR_OPERATION_NOT_INITIALIZED

You might see this error when you use the TPM2 library when running Amazon IoT Greengrass Core as a system service.

This error indicates that you need to add an environment variable that provides the location of the PKCS#11 store in the Amazon IoT Greengrass Core systemd service file.

For more information, see the Requirements section of the PKCS#11 provider component documentation.

Greengrass core device stuck on nucleus v2.12.3

If your Greengrass core device won't revise your deployment from nucleus version 2.12.3, you might need to download and replace the Greengrass.jar file with Greengrass nucleus version 2.12.2. Do the following:

Amazon IoT Greengrass cloud issues

Use the following information to troubleshoot issues with the Amazon IoT Greengrass console and API. Each entry corresponds to an error message that you might see when you perform an action.

An error occurred (AccessDeniedException) when calling the CreateComponentVersion operation: User: arn:aws-cn:iam::123456789012:user/<username> is not authorized to perform: null

You might see this error when you create a component version from the Amazon IoT Greengrass console or with the CreateComponentVersion operation.

This error indicates that your recipe isn't valid JSON or YAML. Check the syntax of your recipe, fix any syntax issues, and try again. You can use an online JSON or YAML syntax checker to identify syntax issues in your recipe.

Invalid Input: Encountered following errors in Artifacts: {<s3ArtifactUri> = Specified artifact resource cannot be accessed}

You might see this error when you create a component version from the Amazon IoT Greengrass console or with the CreateComponentVersion operation. This error indicates that an S3 artifact in the component recipe isn't valid.

Do the following:

INACTIVE deployment status

You might get an INACTIVE deployment status when you call the ListDeployments API without the required dependent Amazon IoT policies. You must have the necessary permissions in order to get an accurate deployment status. You can find the dependent actions by looking in the Actions defined by Amazon IoT Greengrass V2 and following the permissions needed for ListDeployments. Without the required dependent Amazon IoT permissions, you will still see the deployment status but you might see an inaccurate deployment status of INACTIVE.

Core device deployment issues

Troubleshoot deployment issues on Greengrass core devices. Each entry corresponds to a log message that you might see on your core device.

Error: com.aws.greengrass.componentmanager.exceptions.PackageDownloadException: Failed to download artifact

You might see this error when the Amazon IoT Greengrass Core software fails to download a component artifact when the core device applies a deployment. The deployment fails as a result of this error.

When you receive this error, the log also includes a stack trace that you can use to identify the specific issue. Each of the following entries corresponds to a message that you might see in the stack trace of the Failed to download artifact error message.

software.amazon.awssdk.services.s3.model.S3Exception: null (Service: S3, Status Code: 403, Request ID: null, ...)

The PackageDownloadException error might include this stack trace in the following cases:

software.amazon.awssdk.services.s3.model.S3Exception: Access Denied (Service: S3, Status Code: 403, Request ID: <requestID>

The PackageDownloadException error might include this stack trace when the core device doesn't have permission to call s3:GetBucketLocation. The error message also includes the following message.

reason: Failed to determine S3 bucket location

Check that the core device's token exchange role allows s3:GetBucketLocation for the S3 bucket where the artifact is available.

Error: com.aws.greengrass.componentmanager.exceptions.ArtifactChecksumMismatchException: Integrity check for downloaded artifact failed. Probably due to file corruption.

You might see this error when the Amazon IoT Greengrass Core software fails to download a component artifact when the core device applies a deployment. The deployment fails because the downloaded artifact file's checksum doesn't match the checksum that Amazon IoT Greengrass calculated when you created the component.

Do the following:

Error: com.aws.greengrass.componentmanager.exceptions.NoAvailableComponentVersionException: Failed to negotiate component <name> version with cloud and no local applicable version satisfying requirement <requirements>

You might see this error when a core device can't find a component version that meets the requirements of the deployments for that core device. The core device checks for the component in the Amazon IoT Greengrass service and on the local device. The error message includes each deployment's target and that deployment's version requirements for the component. The deployment target can be a thing, a thing group, or LOCAL_DEPLOYMENT, which represents the local deployment on the core device.

This issue can occur in the following cases:

To resolve this issue, revise the deployments to include compatible component versions or remove incompatible ones. For more information about how to revise cloud deployments, see Revise deployments. For more information about how to revise local deployments, see the Amazon IoT Greengrass CLI deployment create command.

software.amazon.awssdk.services.greengrassv2data.model.ResourceNotFoundException: The latest version of Component <componentName> doesn't claim platform <coreDevicePlatform> compatibility

You might see this error when you deploy a component to a core device, and the component doesn't list a platform that is compatible with the core device's platform. Do one of the following:

com.aws.greengrass.componentmanager.exceptions.PackagingException: The deployment attempts to update the nucleus from aws.greengrass.Nucleus-<version> to aws.greengrass.Nucleus-<version> but no component of type nucleus was included as target component

You might see this error when you deploy a component that depends on the Greengrass nucleus, and the core device runs an earlier Greengrass nucleus version than the latest minor version available. This error occurs because the Amazon IoT Greengrass Core software tries to automatically update components to the latest compatible version. However, the Amazon IoT Greengrass Core software prevents the Greengrass nucleus from updating to a new minor version, because several Amazon-provided components depend on specific minor versions of the Greengrass nucleus. For more information, see Greengrass nucleus update behavior.

You must revise the deployment to specify the Greengrass nucleus version that you want to use. Do one of the following:

Error: com.aws.greengrass.deployment.exceptions.DeploymentException: Unable to process deployment. Greengrass launch directory is not set up or Greengrass is not set up as a system service

You might see this error when you move a Greengrass device from one thing group to another, and then back to the original group with deployments that require Greengrass to restart.

To resolve this issue, recreate the launch directory for the device. We also strongly recommend upgrading to version 2.9.6 or later of the Greengrass nucleus.

The following is a Linux script to recreate the launch directory. Save the script in a file called fix_directory.sh.

#!/bin/bash

set -e

GG_ROOT=$1
GG_VERSION=$2

CURRENT="$GG_ROOT/alts/current"

if [ ! -L "$CURRENT" ]; then
  mkdir -p $GG_ROOT/alts/directory_fix
  echo "Relinking $GG_ROOT/alts/directory_fix to $CURRENT"
  ln -sf $GG_ROOT/alts/directory_fix $CURRENT
fi

TARGET=$(readlink $CURRENT)

if [[ ! -d "$TARGET" ]]; then
  echo "Creating directory: $TARGET"
  mkdir -p "$TARGET"
fi

DISTRO_LINK="$TARGET/distro"
DISTRO="$GG_ROOT/packages/artifacts-unarchived/aws.greengrass.Nucleus/$GG_VERSION/aws.greengrass.nucleus/"
echo "Relinking Nucleus artifacts to $DISTRO_LINK"
ln -sf $DISTRO $DISTRO_LINK

To run the script, execute the following command:

[root@ip-172-31-27-165 ~]# ./fix_directory.sh /greengrass/v2 2.9.5
Relinking /greengrass/v2/alts/directory_fix to /greengrass/v2/alts/current
Relinking Nucleus artifacts to /greengrass/v2/alts/directory_fix/distro

Info: com.aws.greengrass.deployment.exceptions.RetryableDeploymentDocumentDownloadException: Greengrass Cloud Service returned an error when getting full deployment configuration

You might see this error when the core device receives a large deployment document, which is a deployment document larger than 7 KB (for deployments that target things) or 31 KB (for deployments that target thing groups). To retrieve a large deployment document, a core device's Amazon IoT policy must allow the greengrass:GetDeploymentConfiguration permission. This error can occur when the core device doesn't have this permission. When this error occurs, the deployment retries indefinitely, and its status is In progress (IN_PROGRESS).

To resolve this issue, add the greengrass:GetDeploymentConfiguration permission to the core device's Amazon IoT policy. For more information, see Update a core device's Amazon IoT policy.

Warn: com.aws.greengrass.deployment.DeploymentService: Failed to get thing group hierarchy

You might see this warning when the core device receives a deployment and the core device's Amazon IoT policy doesn't allow the greengrass:ListThingGroupsForCoreDevice permission. When you create a deployment, the core device uses this permission to identify its thing groups and remove components for any thing groups from which you removed the core device. If the core device runs Greengrass nucleus v2.5.0, the deployment fails. If the core device runs Greengrass nucleus v2.5.1 or later, the deployment proceeds but doesn't remove components. For more information about thing group removal behavior, see Deploy Amazon IoT Greengrass components to devices.

To update the core device's behavior to remove components for thing groups from which you remove the core device, add the greengrass:ListThingGroupsForCoreDevice permission to the core device's Amazon IoT policy. For more information, see Update a core device's Amazon IoT policy.

Info: com.aws.greengrass.deployment.DeploymentDocumentDownloader: Calling Greengrass cloud to get full deployment configuration

You might see this information message printed multiple times without an error, because the core device logs the error at the DEBUG log level. This issue can occur when the core device receives a large deployment document. When this issue occurs, the deployment retries indefinitely, and its status is In progress (IN_PROGRESS). For more information about how to resolve this issue, see this troubleshooting entry.

Caused by: software.amazon.awssdk.services.greengrassv2data.model.GreengrassV2DataException: null (Service: GreengrassV2Data, Status Code: 403, Request ID: <some_request_id>, Extended Request ID: null)

You might see this error when a dataplane API doesn't have iot:Connect permission. If you don't have the correct policy, you'll receive a GreengrassV2DataException: 403. To create a permission policy, follow these instructions: Create an Amazon IoT policy.

Core device component issues

Troubleshoot Greengrass component issues on core devices.

Warn: '<command>' is not recognized as an internal or external command

You might see this error in a Greengrass component's logs when the Amazon IoT Greengrass Core software fails to run a command in the component's lifecycle script. The component's state becomes BROKEN as a result of this error. This error can occur if the system user that runs the component, such as ggc_user, can't find the command's executable in the folders in the PATH.

On Windows devices, check that the folder that contains the executable is in the PATH for the system user that runs the component. If it's missing from the PATH, do one of the following:

Python script doesn't log messages

Greengrass core devices collect logs that you can use to identify issues with components. If your Python script's stdout and stderr messages don't appear in your component logs, you might need to flush the buffer or disable buffering for these standard output streams in Python. Do any of the following:

For more information about how to verify that your Python script outputs log messages, see Monitor Amazon IoT Greengrass logs.

Component configuration doesn't update when changing default configuration

When you change the DefaultConfiguration in a component's recipe, the new default configuration won't replace the component's existing configuration during a deployment. To apply the new default configuration, you must reset the component's configuration to its default settings. When you deploy the component, specify a single empty string as the reset update.

awsiot.greengrasscoreipc.model.UnauthorizedError

You might see this error in a Greengrass component's logs when the component doesn't have permission to perform an IPC operation on a resource. To grant a component permission to call an IPC operation, define an IPC authorization policy in the component's configuration. For more information, see Authorize components to perform IPC operations.

com.aws.greengrass.authorization.exceptions.AuthorizationException: Duplicate policy ID "<id>" for principal "<componentList>"

You might see this error if multiple IPC authorization policies, including across all components on the core device, use the same policy ID.

Check your components' IPC authorization policies, fix any duplicates, and try again. To create unique policy IDs, we recommend that you combine the component name, IPC service name, and a counter. For more information, see Authorize components to perform IPC operations.

com.aws.greengrass.tes.CredentialRequestHandler: Error in retrieving AwsCredentials from TES (HTTP 400)

You might see this error when a core device can't get Amazon credentials from the token exchange service. The HTTP 400 status code indicates that this error occurred because the core device's token exchange IAM role doesn't exist or doesn't have a trust relationship that allows the Amazon IoT credentials provider to assume it.

Do the following:

com.aws.greengrass.tes.CredentialRequestHandler: Error in retrieving AwsCredentials from TES (HTTP 403)

You might see this error when a core device can't get Amazon credentials from the token exchange service. The HTTP 403 status code indicates that this error occurred because the core device's Amazon IoT policies don't grant the iot:AssumeRoleWithCertificate permission for the core device's Amazon IoT role alias.

Review the core device's Amazon IoT policies, and add the iot:AssumeRoleWithCertificate permission for the core device's Amazon IoT role alias. The error message includes the core device's current Amazon IoT role alias. For more information about this permission and how to update the core device's Amazon IoT policies, see Minimal Amazon IoT policy for Amazon IoT Greengrass V2 core devices and Update a core device's Amazon IoT policy.

com.aws.greengrass.tes.CredentialsProviderError: Could not load credentials from any providers

You might see this error when the component tries to request Amazon credentials and can't connect to the token exchange service.

Do the following:

Received error when attempting to retrieve ECS metadata: Could not connect to the endpoint URL: "<tokenExchangeServiceEndpoint>"

You might see this error when the component doesn't run the token exchange service and a component tries to request Amazon credentials.

Do the following:

copyFrom: <configurationPath> is already a container, not a leaf

You might see this error when you change a configuration value from a container type (a list or object) to a non-container type (a string, number, or Boolean). Do the following:

Then, you can set that configuration value to a string, number, or Boolean.

com.aws.greengrass.componentmanager.plugins.docker.exceptions.DockerLoginException: Error logging into the registry using credentials - 'The stub received bad data.'

You might see this error in the Greengrass nucleus logs when the Docker application manager component tries to download a Docker image from a private repository in Amazon Elastic Container Registry (Amazon ECR). This error occurs if you use the wincred Docker credential helper (docker-credential-wincred). As a result, Amazon ECR is unable to store the login credentials.

Take one of the following actions:

java.io.IOException: Cannot run program "cmd" ...: [LogonUser] The password for this account has expired.

You might see this error on a Windows core device when the system user that runs the component's processes, such as ggc_user, has an expired password. As a result, the Amazon IoT Greengrass Core software is unable to run component processes as that system user.

aws.greengrass.StreamManager: Instant exceeds minimum or maximum instant

When you upgrade stream manager v2.0.7 to a version between v2.0.8 and v2.0.11, you might see the following error in the stream manager component's logs if the component fails to start.

2021-07-16T00:54:58.568Z [INFO] (Copier) aws.greengrass.StreamManager: stdout. Caused by: com.fasterxml.jackson.databind.JsonMappingException: Instant exceeds minimum or maximum instant (through reference chain: com.amazonaws.iot.greengrass.streammanager.export.PersistedSuccessExportStatesV1["lastExportTime"]). {scriptName=services.aws.greengrass.StreamManager.lifecycle.startup.script, serviceName=aws.greengrass.StreamManager, currentState=STARTING}
2021-07-16T00:54:58.579Z [INFO] (Copier) aws.greengrass.StreamManager: stdout. Caused by: java.time.DateTimeException: Instant exceeds minimum or maximum instant. {scriptName=services.aws.greengrass.StreamManager.lifecycle.startup.script, serviceName=aws.greengrass.StreamManager, currentState=STARTING}

If you deployed stream manager v2.0.7 and you want to upgrade to a later version, you must upgrade to stream manager v2.0.12 directly. For more information about the stream manager component, see Stream manager.

Core device Lambda function component issues

Troubleshoot Lambda function component issues on core devices.

The following cgroup subsystems are not mounted: devices, memory

You might see this error when you run a containerized Lambda function in the following cases:

To enable cgroups v1, boot the device with the following Linux kernel parameters.

cgroup_enable=memory cgroup_memory=1 systemd.unified_cgroup_hierarchy=0

ipc_client.py:64,HTTP Error 400:Bad Request, b'No subscription exists for the source <label-or-lambda-arn> and subject <label-or-lambda-arn>

You might see this error when you run a V1 Lambda function, which uses the Amazon IoT Greengrass Core SDK, on a V2 core device without specifying a subscription in the legacy subscription router component. To fix this issue, deploy and configure the legacy subscription router to specify the required subscriptions. For more information, see Import V1 Lambda functions.

Component version discontinued

You might see a notification on your Personal Health Dashboard (PHD) when a component version on your core device is discontinued. The component version sends this notification to your PHD within 60 minutes of being discontinued.

To see which deployments you need to revise, do the following using the Amazon Command Line Interface:

Greengrass Command Line Interface issues

Troubleshoot issues with the Greengrass CLI.

java.lang.RuntimeException: Unable to create ipc client

You might see this error when you run a Greengrass CLI command and you specify a different root folder than where the Amazon IoT Greengrass Core software is installed.

Do one of the following to set the root path, and replace /greengrass/v2 with the path to your Amazon IoT Greengrass Core software installation:

Amazon Command Line Interface issues

Troubleshoot Amazon CLI issues for Amazon IoT Greengrass V2.

Error: Invalid choice: 'greengrassv2'

You might see this error when you run an Amazon IoT Greengrass V2 command with the Amazon CLI (for example, aws greengrassv2 list-core-devices).

This error indicates that you have a version of the Amazon CLI that doesn't support Amazon IoT Greengrass V2. To use Amazon IoT Greengrass V2 with the Amazon CLI, you must have one of the following versions or later:

aws --version

To resolve this issue, update the Amazon CLI to a later version that supports Amazon IoT Greengrass V2. For more information, see Installing, updating, and uninstalling the Amazon CLI in the Amazon Command Line Interface User Guide.