DynamoDB EPROTO Error

[cmoyer2]

Describe the bug

Sporadically receiving the following error when trying to connect to DynamoDB from various different Lambda functions. This error does not consistently occur in any one particular Lambda function. We see this error occur in different Lambdas in our environment, but when it does occur it always in a Lambda attempting to connect to DynamoDB.

{
      "errorType": "Error",
      "errorMessage": "write EPROTO",
      "code": "EPROTO",
      "errno": -71,
      "syscall": "write",
      "$metadata": {
          "attempts": 1,
          "totalRetryDelay": 0
}

Your environment

SDK version number

"@aws-sdk/client-dynamodb": "^3.49.0",
"@aws-sdk/lib-dynamodb": "^3.49.0",

Is the issue in the browser/Node.js/ReactNative?

Node.js

Details of the browser/Node.js/ReactNative version

Lambda Node.js 14.x Runtime

Steps to reproduce

Cannot consistently reproduce, error occurs roughly once a week in our serverless environment. We have multiple Lambdas that access DynamoDB, and not all of them fail at the same time, but we have seen this error occur in all Lambdas that do access DynamoDB.

We use the following code to create the DynamoDB client. We then import the documentClient into the handler for each Lambda that needs access to DynamoDB.

import {DynamoDBDocumentClient} from "@aws-sdk/lib-dynamodb";
import {DynamoDBClient} from "@aws-sdk/client-dynamodb";

const dynamoDBClient = new DynamoDBClient({

});

const marshallOptions = {
  // Whether to automatically convert empty strings, blobs, and sets to `null`.
  convertEmptyValues: false, // false, by default.
  // Whether to remove undefined values while marshalling.
  removeUndefinedValues: true, // false, by default.
  // Whether to convert typeof object to map attribute.
  convertClassInstanceToMap: false, // false, by default.
};

const unmarshallOptions = {
  // Whether to return numbers as a string instead of converting them to native JavaScript numbers.
  wrapNumbers: false, // false, by default.
};

const translateConfig = { marshallOptions, unmarshallOptions };

export const documentClient = DynamoDBDocumentClient.from(dynamoDBClient, translateConfig);

Observed behavior

Receive the following error when trying to access DynamoDB with documentClient.send() with a GetItemCommand, QueryCommand, UpdateItemCommand, or PutItemCommand.

{
    "errorType": "Runtime.UnhandledPromiseRejection",
    "errorMessage": "Error: write EPROTO",
    "reason": {
        "errorType": "Error",
        "errorMessage": "write EPROTO",
        "code": "EPROTO",
        "errno": -71,
        "syscall": "write",
        "$metadata": {
            "attempts": 1,
            "totalRetryDelay": 0
        },
        "stack": [
            "Error: write EPROTO",
            "    at WriteWrap.onWriteComplete [as oncomplete] (internal/stream_base_commons.js:94:16)"
        ]
    },
    "promise": {},
    "stack": [
        "Runtime.UnhandledPromiseRejection: Error: write EPROTO",
        "    at process.<anonymous> (/var/runtime/index.js:35:15)",
        "    at process.emit (events.js:400:28)",
        "    at processPromiseRejections (internal/process/promises.js:245:33)",
        "    at processTicksAndRejections (internal/process/task_queues.js:96:32)"
    ]
}

Expected behavior

DocumentClient consistently able to access DynamoDB without throwing networking errors.

Screenshots

If applicable, add screenshots to help explain your problem.

Additional context

I've viewed the following issues that have been previously opened that seem to be related to this issue, although most seem to be around 2016 in much earlier versions of the SDK and Node.js. Most recommendations are to force the secure protocol to TLSv1, but I don't imagine that would be the solution any longer as TLSv1.2 would be the recommended version.

aws/aws-sdk-js#862
nodejs/node#3692

[sutts101]

We have also been seeing the same issue very sporadically. Tentatively it looks like upgrading to ^3.58.0 might have fixed it...

[spidermonk]

I started seeing this issue this past week and just updated to 3.99.0 (Node v14.19.3) but still see the error.

EDITED to add: Over the past 24 hours this happened 17 times from just under 4,000 lambda invocations.

Error:

{
    "errorType": "Error",
    "errorMessage": "write EPROTO",
    "code": "EPROTO",
    "errno": -71,
    "syscall": "write",
    "$metadata": {
        "attempts": 1,
        "totalRetryDelay": 0
    },
    "level": "warn",
    "stack": [
        "Error: write EPROTO",
        "    at WriteWrap.onWriteComplete [as oncomplete] (internal/stream_base_commons.js:94:16)"
    ]
}

Client creation code:

dynamoDocClient = DynamoDBDocument.from(new DynamoDB({region: region}), 
    {
        marshallOptions: {
            removeUndefinedValues: true,
            convertEmptyValues: false
        }
    });

Note: This is from a Lambda to DDB via a VPC endpoint

[vvo]

Hey there, also receiving this error, although I am using the @aws-sdk/client-cloudwatch in my lambda.
Version: "@aws-sdk/client-cloudwatch": "3.118.0",

Error is:

{
  "errorType":"Runtime.UnhandledPromiseRejection",
  "errorMessage":"Error: write EPROTO",
  "reason":{
    "errorType":"Error",
    "errorMessage":"write EPROTO",
    "code":"EPROTO",
    "errno":-71,
    "syscall":"write",
    "$metadata":{
      "attempts":1,
      "totalRetryDelay":0
    },
    "stack":[
      "Error: write EPROTO",
      "    at WriteWrap.onWriteComplete [as oncomplete] (internal/stream_base_commons.js:94:16)"
    ]
  },
  "promise":{
    
  },
  "stack":[
    "Runtime.UnhandledPromiseRejection: Error: write EPROTO",
    "    at process.<anonymous> (/var/runtime/index.js:35:15)",
    "    at process.emit (events.js:412:35)",
    "    at process.emit (/var/task/node_modules/source-map-support/source-map-support.js:516:21)",
    "    at processPromiseRejections (internal/process/promises.js:245:33)",
    "    at processTicksAndRejections (internal/process/task_queues.js:96:32)"
  ]
}

[sirimykland]

I also experience this exact error sporadically in our lambdas running node 14. It somewhat seems to relate to dynamodb, as it only occur in the lambdas that use a dynamodb connection. I tried to upgrade @aws-sdk/* to the latest version 3.121.0 without any luck. Any help would be greatly appreciated!

[jcardus]

This happens just with require('aws-sdk/clients/dynamodb'), even if you don't do anything...

[RanVaknin]

Hi there,

Im sorry for the long wait on the thread. This is something I have not seen come up in a while, and my guess is that it got resolved in newer versions.
Can someone on this thread install the latest version and see if this persists?

Thanks,
Ran~

[github-actions]

This issue has not received a response in 1 week. If you still think there is a problem, please leave a comment to avoid the issue from automatically closing.

[github-actions]

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.