Pagination: unknown error, exception or failure

[arivera-tiempodev]

I have a DynamoDB table that is generated with the following code:

const params = {
  TableName,
  AttributeDefinitions: [{
    AttributeName: 'post_id',
    AttributeType: 'S'
  }, {
    AttributeName: 'parent_post_id',
    AttributeType: 'S'
  }, {
    AttributeName: 'created_at',
    AttributeType: 'S'
  }, {
    AttributeName: 'average_rating',
    AttributeType: 'N'
  }],
  KeySchema: [{
    AttributeName: 'post_id',
    KeyType: 'HASH'
  }, {
    AttributeName: 'created_at',
    KeyType: 'RANGE'
  }],
  GlobalSecondaryIndexes: [{
    IndexName: 'parent_post_id-created_at-index',
    ProvisionedThroughput: {
      ReadCapacityUnits: 5,
      WriteCapacityUnits: 5
    },
    KeySchema: [{
      AttributeName: 'parent_post_id',
      KeyType: 'HASH'
    }, {
      AttributeName: 'created_at',
      KeyType: 'SORT'
    }],
    Projection: {
      ProjectionType: 'ALL'
    }
  }, {
    IndexName: 'parent_post_id-averge_rating-index',
    ProvisionedThroughput: {
      ReadCapacityUnits: 5,
      WriteCapacityUnits: 5
    },
    KeySchema: [{
      AttributeName: 'parent_post_id',
      KeyType: 'HASH'
    }, {
      AttributeName: 'average_rating',
      KeyType: 'SORT'
    }],
    Projection: {
      ProjectionType: 'ALL'
    }
  }],
  ProvisionedThroughput: {
    ReadCapacityUnits: 1,
    WriteCapacityUnits: 1
  }
};

Then I make queries like this:

const paramsPosts = {
  TableName: this.table,
  IndexName: 'parent_post_id-created_at-index',
  KeyConditionExpression: '#parent_post_id = :parent_post_id',
  FilterExpression: '#resource_id = :resource_id',
  ExpressionAttributeNames: {
    '#parent_post_id': 'parent_post_id',
    '#resource_id': 'resource_id',
  },
  ExpressionAttributeValues: {
    ':parent_post_id': '-',
    ':resource_id': resource_id,
  },
  Limit: limit,
  ScanIndexForward: true,
  ExclusiveStartKey: lastEvaluatedKey
};

When I send null or undefined as LastEvaluatedKey I get expected results:

{
  "resource_id": "VALID_RESOURCE_ID",
  "limit": 5
}

RESULT:

{
  "posts": {
    "items": [
      {
        "comment": "This comment 1.",
        "created_at": "2018-07-31T19:40:21.899Z"
      },
      {
        "comment": "This comment 2.",
        "created_at": "2018-07-31T19:40:26.460Z"
      },
      {
        "comment": "This comment 3.",
        "created_at": "2018-07-31T19:40:30.810Z"
      },
      {
        "comment": "This comment 4.",
        "created_at": "2018-07-31T19:40:35.057Z"
      },
      {
        "comment": "This comment 5.",
        "created_at": "2018-07-31T19:40:39.341Z"
      }
    ],
      "summary": {
      "total_count": 13
    },
    "paging": {
      "next": {
        "post_id": "660b4878-e883-41ce-af4e-c393a20aac08",
        "created_at": "2018-07-31T19:40:39.341Z",
        "parent_post_id": "-"
      },
      "prev": null
    }
  }
}

I got that object because I formatting like this:

const response = {
  items: [],
  paging: {},
  summary: {}
};

return this.query(paramsPosts)
  .then(postsData => {
    response.paging.next = postsData.LastEvaluatedKey;
    response.items = postsData.Items;
    return this.query(paramsCount);
  }).then(countData => {
    response.paging.prev = lastEvaluatedKey;
    response.summary.total_count = countData.Count;
    return response;
  });

Then, if I use posts.paging.next as ExclusiveStartKey I got the following error:

{
  "data": {
    "posts": null
  },
  "errors": [
    {
      "message": "The request processing has failed because of an unknown error, exception or failure.",
      "locations": [
        {
          "line": 2,
          "column": 3
        }
      ],
      "path": [
        "posts"
      ],
      "extensions": {
        "code": "INTERNAL_SERVER_ERROR",
        "exception": {
          "message": "The request processing has failed because of an unknown error, exception or failure.",
          "code": "InternalFailure",
          "time": "2018-07-31T21:16:56.008Z",
          "requestId": "10dc5a07-a18e-407b-bba3-ad57506d5fd2",
          "statusCode": 500,
          "retryable": true,
          "stacktrace": [
            "InternalFailure: The request processing has failed because of an unknown error, exception or failure.",
            "    at Request.extractError (/home/node/node_modules/aws-sdk/lib/protocol/json.js:48:27)",
            "    at Request.callListeners (/home/node/node_modules/aws-sdk/lib/sequential_executor.js:105:20)",
            "    at Request.emit (/home/node/node_modules/aws-sdk/lib/sequential_executor.js:77:10)",
            "    at Request.emit (/home/node/node_modules/aws-sdk/lib/request.js:683:14)",
            "    at Request.transition (/home/node/node_modules/aws-sdk/lib/request.js:22:10)",
            "    at AcceptorStateMachine.runTo (/home/node/node_modules/aws-sdk/lib/state_machine.js:14:12)",
            "    at /home/node/node_modules/aws-sdk/lib/state_machine.js:26:10",
            "    at Request.<anonymous> (/home/node/node_modules/aws-sdk/lib/request.js:38:9)",
            "    at Request.<anonymous> (/home/node/node_modules/aws-sdk/lib/request.js:685:12)",
            "    at Request.callListeners (/home/node/node_modules/aws-sdk/lib/sequential_executor.js:115:18)"
          ]
        }
      }
    }
  ]
}

I have DynamoDB running in local inside of a Docker container with the InMemory flag set to true.

I thought that I only have to send GSI Key, that is composed by parent_post_id (partition key) and created_at (sort key) but reading in some places I found that I also have to send Table Key. If I remove post_id that is only part of Table Key I got the following error:

{
  "data": {
    "posts": null
  },
  "errors": [
    {
      "message": "Exclusive Start Key must have same size as table's key schema",
      "locations": [
        {
          "line": 2,
          "column": 3
        }
      ],
      "path": [
        "posts"
      ],
      "extensions": {
        "code": "INTERNAL_SERVER_ERROR",
        "exception": {
          "message": "Exclusive Start Key must have same size as table's key schema",
          "code": "ValidationException",
          "time": "2018-07-31T21:20:11.286Z",
          "requestId": "f8961dcd-325c-4cda-8fa8-855539cae715",
          "statusCode": 400,
          "retryable": false,
          "stacktrace": [
            "ValidationException: Exclusive Start Key must have same size as table's key schema",
            "    at Request.extractError (/home/node/node_modules/aws-sdk/lib/protocol/json.js:48:27)",
            "    at Request.callListeners (/home/node/node_modules/aws-sdk/lib/sequential_executor.js:105:20)",
            "    at Request.emit (/home/node/node_modules/aws-sdk/lib/sequential_executor.js:77:10)",
            "    at Request.emit (/home/node/node_modules/aws-sdk/lib/request.js:683:14)",
            "    at Request.transition (/home/node/node_modules/aws-sdk/lib/request.js:22:10)",
            "    at AcceptorStateMachine.runTo (/home/node/node_modules/aws-sdk/lib/state_machine.js:14:12)",
            "    at /home/node/node_modules/aws-sdk/lib/state_machine.js:26:10",
            "    at Request.<anonymous> (/home/node/node_modules/aws-sdk/lib/request.js:38:9)",
            "    at Request.<anonymous> (/home/node/node_modules/aws-sdk/lib/request.js:685:12)",
            "    at Request.callListeners (/home/node/node_modules/aws-sdk/lib/sequential_executor.js:115:18)"
          ]
        }
      }
    }
  ]
}

And DynamoDB print this on console:

dynamo-forum | Jul 31, 2018 9:32:31 PM com.almworks.sqlite4java.Internal log
dynamo-forum | WARNING: [sqlite] SQLiteDBAccess$20@268969a1: job exception
dynamo-forum | java.lang.NullPointerException
dynamo-forum |  at com.amazonaws.services.dynamodbv2.local.shared.access.sqlite.SQLiteDBAccessUtils.translateKeyAttributeValue(SQLiteDBAccessUtils.java:194)
dynamo-forum |  at com.amazonaws.services.dynamodbv2.local.shared.access.sqlite.SQLiteDBAccess$20.doWork(SQLiteDBAccess.java:2012)
dynamo-forum |  at com.amazonaws.services.dynamodbv2.local.shared.access.sqlite.SQLiteDBAccess$20.doWork(SQLiteDBAccess.java:1787)
dynamo-forum |  at com.amazonaws.services.dynamodbv2.local.shared.access.sqlite.AmazonDynamoDBOfflineSQLiteJob.job(AmazonDynamoDBOfflineSQLiteJob.java:97)
dynamo-forum |  at com.almworks.sqlite4java.SQLiteJob.execute(SQLiteJob.java:372)
dynamo-forum |  at com.almworks.sqlite4java.SQLiteQueue.executeJob(SQLiteQueue.java:534)
dynamo-forum |  at com.almworks.sqlite4java.SQLiteQueue.queueFunction(SQLiteQueue.java:667)
dynamo-forum |  at com.almworks.sqlite4java.SQLiteQueue.runQueue(SQLiteQueue.java:623)
dynamo-forum |  at com.almworks.sqlite4java.SQLiteQueue.access$000(SQLiteQueue.java:77)
dynamo-forum |  at com.almworks.sqlite4java.SQLiteQueue$1.run(SQLiteQueue.java:205)
dynamo-forum |  at java.lang.Thread.run(Thread.java:748)
dynamo-forum |

And this is by AWS-SDK NPM package:

service-forum | [AWS dynamodb 500 0.025s 0 retries] query({ TableName: 'forum-post',
service-forum |   IndexName: 'parent_post_id-created_at-index',
service-forum |   KeyConditionExpression: '#parent_post_id = :parent_post_id',
service-forum |   FilterExpression: '#resource_id = :resource_id',
service-forum |   ExpressionAttributeNames:
service-forum |    { '#parent_post_id': 'parent_post_id',
service-forum |      '#resource_id': 'resource_id' },
service-forum |   ExpressionAttributeValues:
service-forum |    { ':parent_post_id': { S: '-' },
service-forum |      ':resource_id': { S: 'VALID_RESOURCE_ID' } },
service-forum |   Limit: 5,
service-forum |   ScanIndexForward: true,
service-forum |   ExclusiveStartKey:
service-forum |    { post_id: { S: '660b4878-e883-41ce-af4e-c393a20aac08' },
service-forum |      created_at: { S: '2018-07-31T19:40:39.341Z' },
service-forum |      parent_post_id: { S: '-' } } 

[AllanZhengYP]

Hi @arivera-tiempodev Thank you for reporting this issue. I can reproduce it and it seems that this issue only happens in local DynamoDB and it works fine remotely. I'm not sure if this is a known issue, I will contact the service team to verify. You can also call out at the DynamoDB forum. One thing is sure that this is not an SDK issue.

I will comment here if hear back from the service team.

[ajredniwja]

Closing the issue now since the fix is being tracked by service team in this PR.

Please reach out if you have any questions.

[lock]

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs and link to relevant comments in this thread.