Lambda Function timeout error

Question

Hi AWS, I am writing a lambda function where it will list all the S3 buckets, list the tag with keyword `cost` in the key and fix it if there is a typo or semantics mistake, for e.g. cost_centre, cost_Centre, cost_centre & cost_Center to **cost_centre**. I have written this code for the same.
```
import json
import boto3
from botocore.exceptions import ClientError

def lambda_handler(event, context):
    client = boto3.client(service_name='s3')
    response = client.list_buckets()

# list s3 buckets and the tags associated with the bucket
    for bucket in response['Buckets']:
        try:
            result = client.get_bucket_tagging(
                Bucket=bucket['Name']
                )
            value = result['TagSet']
            for tags in value:
                if 'cost' in tags['Key']:
                    print(f"The bucket name is: {bucket['Name']} and the tags are: {tags['Key']}:{tags['Value']}")
        except ClientError as e:
            if e.response['Error']['Code'] == 'TagSetNotFoundError':
                print("TagSet Not Found Error" + e)
            else:
                print(f"The bucket name is: {bucket['Name']} has no tags")

# replace wrong tag with the right tag key
    for bucket in response['Buckets']:
        if tags['Key'] in ('cost_center', 'costcenter', 'costcentre', 'cost_Center', 'cost_Centre'):
            update_tag = client.put_bucket_tagging(
                Bucket=bucket['Name'],
                Tagging={
                    'TagSet': [
                        {
                            'Key': 'cost_centre',
                            'Value': tags['Value']
                        }
                    ]
                }
            )

```

We have more than 50 buckets in a given AWS account but instead of evaluating the results within the default time (15 mins) it is giving timeout. Can you help where I need to refactor my code?

**NOTE**: Each bucket has 5-6 tags like primary_owner, secondardy_owner etc. but I am only interested in checking **cost_centre**

Thanks

Accepted Answer

Final working code:

```
import boto3
from botocore.exceptions import ClientError

# Tag with incorrect semantics/name
incorrect_tag = ['cost_center', 'cost_Center', 'cost_Centre', 'costcentre', 'costcenter', 'costCentre', 'costCenter', 'CostCenter', 'CostCentre', 'Cost_centre', 'Cost_center']

def lambda_handler(event, context):
    client = boto3.client(service_name='s3')
    response = client.list_buckets()

# list s3 buckets and the tags associated with the bucket
    for bucket in response['Buckets']:
        try:
            result = client.get_bucket_tagging(
                Bucket=bucket['Name']
                )
            values = result['TagSet']
            for tags in values:
                if tags['Key'] in incorrect_tag:
                    tags['Key'] = 'cost_centre'
            print(values)
            
            # applying the tags to s3 bucket
            s3_add_tag = client.put_bucket_tagging(
                Bucket=bucket['Name'],
                Tagging={
                    'TagSet': values
                }
            )
            print(f"{bucket['Name']} has right tags {tags['Key']}:{tags['Value']}")
                
            print("success")
        except ClientError as e:
            if e.response['Error']['Code'] == 'TagSetNotFoundError':
                print("TagSet Not Found Error" + e)
            else:
                print(f"The bucket name is: {bucket['Name']} has no tags")
```

Answer

Hi, your code doesn't  have dependencies on the input vars of the lambda handler. So, I would suggest that slightly refactor it to isolate a function looping across your list of buckets. Then, you test this function from your laptop to make sure it works as expected.

When you reach this stage, you can start looking at the Lambda config. 
You most probably have an issue with the Lambda execution role: see https://docs.aws.amazon.com/lambda/latest/dg/lambda-intro-execution-role.html

You need to make sure that the role has the proper credentials to list s3 buckets. See https://docs.aws.amazon.com/AmazonS3/latest/userguide/s3-access-control.html for all possible permissions

Answer

The above configurations as suggested by @Uri and @Didier_AWS were there in place. The issue was because of the proxy.

**For Linux or Mac**:

$ export HTTP_PROXY=http://:
$ export HTTPS_PROXY=http://:

**For Windows**:

set HTTP_PROXY=http://:
set HTTPS_PROXY=http://:

But now I am facing a new issue.  There is an issue in this piece of code as though it is updating the semantics and grammar of cost centre tag with the right semantics but it is removing other tags attached to the function and also adding the tag to the bucket which has no tags and also overriding the value of cost centre with the same value for every s3 bucket in the account.

```
# replace wrong tag with the right tag key
    for bucket in response['Buckets']:
        if tags['Key'] in ('cost_center', 'costcenter', 'costcentre', 'cost_Center', 'cost_Centre'):
            update_tag = client.put_bucket_tagging(
                Bucket=bucket['Name'],
                Tagging={
                    'TagSet': [
                        {
                            'Key': 'cost_centre',
                            'Value': tags['Value']
                        }
                    ]
                }
            )
```

Can someone help in fixing this code where it will only update the cost centre tag with wrong semantics and keep the value as it is as my purpose is to only fix the grammar and semantics of the tag **cost_centre**

Answer

Is your Lambda function attached to a VPC? If it is, you either need to create an S3 VPC endpoint, or better yet, remove it from the VPC, as it doesn't access any VPC resources.

If this is not the case, are you seeing any printouts in CloudWatch Logs?  How long does it take to timeout?

Lambda Function timeout error

Relevant content