Get hands-on experience with 20+ free Google Cloud products and $300 in free credit for new customers.

Vertex AI Endpoint won't deploy - unknown error

Every time I try to deploy any model and endpoint on Vertex AI - standard Llama 3.1 model or a Huggingface embedding model, I get an error saying the endpoint can't be created because of an unknown error. This is happening for every model I deploy through the model garden GUI

 

This is what my logs say each time:
{
"protoPayload": {
"@type": "type.googleapis.com/google.cloud.audit.AuditLog",
"status": {
"code": 13,
"message": "java.util.NoSuchElementException: No value present\n\tat java.base/java.util.Optional.get(Unknown Source)\n\tat com.google.cloud.ai.platform.boq.shared.api.endpoint.CreateEndpointApiHandler.lambda$startCreateEndpointPipeline$31(CreateEndpointApiHandler.java:938)\n\tat com.google.intelligence.cloud.automl.boq.publics.pipeline.fluentapi.TaskFactory.createFromTaskMetadataGroup(TaskFactory.java:107)\n\tat com.google.intelligence.cloud.automl.boq.publics.pipeline.fluentapi.TaskFactory.createFromTaskMetadataGroup(TaskFactory.java:95)\n\tat com.google.intelligence.cloud.automl.boq.publics.pipeline.fluentapi.TaskFactory.createFromTaskMetadataGroup(TaskFactory.java:95)\n\tat com.google.intelligence.cloud.automl.boq.publics.pipeline.fluentapi.TaskFactory.createFromTaskMetadataGroup(TaskFactory.java:95)\n\tat com.google.intelligence.cloud.automl.boq.publics.pipeline.fluentapi.TaskFactory.createFromTaskMetadataGroup(TaskFactory.java:95)\n\tat com.google.intelligence.cloud.automl.boq.publics.pipeline.fluentapi.TaskFactory.create(TaskFactory.java:50)\n\tat com.google.intelligence.cloud.automl.boq.publics.pipeline.fluentapi.AbstractFluentPipelineHolder.buildPipelineWithContext(AbstractFluentPipelineHolder.java:583)\n\tat com.google.cloud.ai.platform.boq.shared.pipeline.fluentpipeline.LroFluentPipeline.buildPipeline(LroFluentPipeline.java:365)\n\tat com.google.cloud.ai.platform.boq.shared.pipeline.fluentpipeline.LroFluentPipeline.lambda$build$2(LroFluentPipeline.java:168)\n\tat com.google.async.promisegraph.ThenPromise.lambda$buildResultFromInputs$0(ThenPromise.java:91)\n\tat com.google.async.promisegraph.PromiseResult.fromSupplier(PromiseResult.java:29)\n\tat com.google.async.promisegraph.ThenPromise.buildResultFromInputs(ThenPromise.java:87)\n\tat com.google.async.promisegraph.AbstractPromise.complete(AbstractPromise.java:210)\n\tat com.google.async.promisegraph.GraphEnvironment.pumpLoop(GraphEnvironment.java:367)\n\tat com.google.async.promisegraph.GraphEnvironment.pumpWithoutBlocking(GraphEnvironment.java:305)\n\tat com.google.async.promisegraph.GraphEnvironment.executeWorkUntilBlockedOrCompleted(GraphEnvironment.java:553)\n\tat java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(Unknown Source)\n\tat java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(Unknown Source)\n\tat com.google.apps.framework.server.AbstractThreadPoolModule$InitializingThreadFactory.lambda$newThread$0(AbstractThreadPoolModule.java:424)\n\tat java.base/java.lang.Thread.run(Unknown Source)\n\tSuppressed: com.google.async.promisegraph.PromiseOriginException\n\t\tat com.google.cloud.ai.platform.boq.shared.pipeline.fluentpipeline.LroFluentPipeline.build(LroFluentPipeline.java:165)\n\tSuppressed: java.util.concurrent.ExecutionException: \n"
},
"authenticationInfo": {
"principalEmail": "XXXX@gmail.com"
},
"requestMetadata": {
"callerIp": "24.62.179.155",
"callerSuppliedUserAgent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/127.0.0.0 Safari/537.36,gzip(gfe),gzip(gfe)",
"requestAttributes": {
"time": "2024-08-12T12:37:26.465940566Z",
"reason": "8uSywAYQGg5Db2xpc2V1bSBGbG93cw",
"auth": {}
},
"destinationAttributes": {}
},
"serviceName": "aiplatform.googleapis.com",
"methodName": "google.cloud.aiplatform.ui.EndpointService.CreateEndpoint",
"authorizationInfo": [
{
"resource": "projects/1044600842123/locations/us-east4",
"permission": "aiplatform.endpoints.create",
"granted": true,
"resourceAttributes": {},
"permissionType": "ADMIN_WRITE"
}
],
"resourceName": "projects/1044600842123/locations/us-east4",
"request": {
"@type": "type.googleapis.com/google.cloud.aiplatform.ui.CreateEndpointRequest",
"endpoint": {
"trafficSplit": {
"0": 100
}
},
"parent": "projects/1044600842123/locations/us-east4"
},
"resourceLocation": {
"currentLocations": [
"us-east4"
]
}
},
"insertId": "nm2n6re1kpjh",
"resource": {
"type": "audited_resource",
"labels": {
"service": "aiplatform.googleapis.com",
"project_id": "hale-aurora-429222-t4",
"method": "google.cloud.aiplatform.ui.EndpointService.CreateEndpoint"
}
},
"timestamp": "2024-08-12T12:37:26.459643373Z",
"severity": "ERROR",
"logName": "projects/hale-aurora-429222-t4/logs/cloudaudit.googleapis.com%2Factivity",
"receiveTimestamp": "2024-08-12T12:37:27.023187794Z"
}

0 2 241
2 REPLIES 2

Hi @raviravi6,

Welcome to Google Cloud Community!

Based on my understanding, you're unable to deploy your model to an endpoint using Model Garden. To isolate if this is only happening to a specific environment, you may try deploying your model using the console or API. You may follow the instructions mentioned in the documentation.

Before deploying your model, please consider the following best practices mentioned here by @lsolatorio that can help avoid deployment issues.

I hope the above information is helpful.

I have tried multiple ways to deploy and get the same error. Something seems wrong on the Google back-end