Tag Archives: actions

Analyze data in Amazon DynamoDB using Amazon SageMaker for real-time prediction

Post Syndicated from YongSeong Lee original https://aws.amazon.com/blogs/big-data/analyze-data-in-amazon-dynamodb-using-amazon-sagemaker-for-real-time-prediction/

Many companies across the globe use Amazon DynamoDB to store and query historical user-interaction data. DynamoDB is a fast NoSQL database used by applications that need consistent, single-digit millisecond latency.

Often, customers want to turn their valuable data in DynamoDB into insights by analyzing a copy of their table stored in Amazon S3. Doing this separates their analytical queries from their low-latency critical paths. This data can be the primary source for understanding customers’ past behavior, predicting future behavior, and generating downstream business value. Customers often turn to DynamoDB because of its great scalability and high availability. After a successful launch, many customers want to use the data in DynamoDB to predict future behaviors or provide personalized recommendations.

DynamoDB is a good fit for low-latency reads and writes, but it’s not practical to scan all data in a DynamoDB database to train a model. In this post, I demonstrate how you can use DynamoDB table data copied to Amazon S3 by AWS Data Pipeline to predict customer behavior. I also demonstrate how you can use this data to provide personalized recommendations for customers using Amazon SageMaker. You can also run ad hoc queries using Amazon Athena against the data. DynamoDB recently released on-demand backups to create full table backups with no performance impact. However, it’s not suitable for our purposes in this post, so I chose AWS Data Pipeline instead to create managed backups are accessible from other services.

To do this, I describe how to read the DynamoDB backup file format in Data Pipeline. I also describe how to convert the objects in S3 to a CSV format that Amazon SageMaker can read. In addition, I show how to schedule regular exports and transformations using Data Pipeline. The sample data used in this post is from Bank Marketing Data Set of UCI.

The solution that I describe provides the following benefits:

  • Separates analytical queries from production traffic on your DynamoDB table, preserving your DynamoDB read capacity units (RCUs) for important production requests
  • Automatically updates your model to get real-time predictions
  • Optimizes for performance (so it doesn’t compete with DynamoDB RCUs after the export) and for cost (using data you already have)
  • Makes it easier for developers of all skill levels to use Amazon SageMaker

All code and data set in this post are available in this .zip file.

Solution architecture

The following diagram shows the overall architecture of the solution.

The steps that data follows through the architecture are as follows:

  1. Data Pipeline regularly copies the full contents of a DynamoDB table as JSON into an S3
  2. Exported JSON files are converted to comma-separated value (CSV) format to use as a data source for Amazon SageMaker.
  3. Amazon SageMaker renews the model artifact and update the endpoint.
  4. The converted CSV is available for ad hoc queries with Amazon Athena.
  5. Data Pipeline controls this flow and repeats the cycle based on the schedule defined by customer requirements.

Building the auto-updating model

This section discusses details about how to read the DynamoDB exported data in Data Pipeline and build automated workflows for real-time prediction with a regularly updated model.

Download sample scripts and data

Before you begin, take the following steps:

  1. Download sample scripts in this .zip file.
  2. Unzip the src.zip file.
  3. Find the automation_script.sh file and edit it for your environment. For example, you need to replace 's3://<your bucket>/<datasource path>/' with your own S3 path to the data source for Amazon ML. In the script, the text enclosed by angle brackets—< and >—should be replaced with your own path.
  4. Upload the json-serde-1.3.6-SNAPSHOT-jar-with-dependencies.jar file to your S3 path so that the ADD jar command in Apache Hive can refer to it.

For this solution, the banking.csv  should be imported into a DynamoDB table.

Export a DynamoDB table

To export the DynamoDB table to S3, open the Data Pipeline console and choose the Export DynamoDB table to S3 template. In this template, Data Pipeline creates an Amazon EMR cluster and performs an export in the EMRActivity activity. Set proper intervals for backups according to your business requirements.

One core node(m3.xlarge) provides the default capacity for the EMR cluster and should be suitable for the solution in this post. Leave the option to resize the cluster before running enabled in the TableBackupActivity activity to let Data Pipeline scale the cluster to match the table size. The process of converting to CSV format and renewing models happens in this EMR cluster.

For a more in-depth look at how to export data from DynamoDB, see Export Data from DynamoDB in the Data Pipeline documentation.

Add the script to an existing pipeline

After you export your DynamoDB table, you add an additional EMR step to EMRActivity by following these steps:

  1. Open the Data Pipeline console and choose the ID for the pipeline that you want to add the script to.
  2. For Actions, choose Edit.
  3. In the editing console, choose the Activities category and add an EMR step using the custom script downloaded in the previous section, as shown below.

Paste the following command into the new step after the data ­­upload step:

s3://#{myDDBRegion}.elasticmapreduce/libs/script-runner/script-runner.jar,s3://<your bucket name>/automation_script.sh,#{output.directoryPath},#{myDDBRegion}

The element #{output.directoryPath} references the S3 path where the data pipeline exports DynamoDB data as JSON. The path should be passed to the script as an argument.

The bash script has two goals, converting data formats and renewing the Amazon SageMaker model. Subsequent sections discuss the contents of the automation script.

Automation script: Convert JSON data to CSV with Hive

We use Apache Hive to transform the data into a new format. The Hive QL script to create an external table and transform the data is included in the custom script that you added to the Data Pipeline definition.

When you run the Hive scripts, do so with the -e option. Also, define the Hive table with the 'org.openx.data.jsonserde.JsonSerDe' row format to parse and read JSON format. The SQL creates a Hive EXTERNAL table, and it reads the DynamoDB backup data on the S3 path passed to it by Data Pipeline.

Note: You should create the table with the “EXTERNAL” keyword to avoid the backup data being accidentally deleted from S3 if you drop the table.

The full automation script for converting follows. Add your own bucket name and data source path in the highlighted areas.

#!/bin/bash
hive -e "
ADD jar s3://<your bucket name>/json-serde-1.3.6-SNAPSHOT-jar-with-dependencies.jar ; 
DROP TABLE IF EXISTS blog_backup_data ;
CREATE EXTERNAL TABLE blog_backup_data (
 customer_id map<string,string>,
 age map<string,string>, job map<string,string>, 
 marital map<string,string>,education map<string,string>, 
 default map<string,string>, housing map<string,string>,
 loan map<string,string>, contact map<string,string>, 
 month map<string,string>, day_of_week map<string,string>, 
 duration map<string,string>, campaign map<string,string>,
 pdays map<string,string>, previous map<string,string>, 
 poutcome map<string,string>, emp_var_rate map<string,string>, 
 cons_price_idx map<string,string>, cons_conf_idx map<string,string>,
 euribor3m map<string,string>, nr_employed map<string,string>, 
 y map<string,string> ) 
ROW FORMAT SERDE 'org.openx.data.jsonserde.JsonSerDe' 
LOCATION '$1/';

INSERT OVERWRITE DIRECTORY 's3://<your bucket name>/<datasource path>/' 
SELECT concat( customer_id['s'],',', 
 age['n'],',', job['s'],',', 
 marital['s'],',', education['s'],',', default['s'],',', 
 housing['s'],',', loan['s'],',', contact['s'],',', 
 month['s'],',', day_of_week['s'],',', duration['n'],',', 
 campaign['n'],',',pdays['n'],',',previous['n'],',', 
 poutcome['s'],',', emp_var_rate['n'],',', cons_price_idx['n'],',',
 cons_conf_idx['n'],',', euribor3m['n'],',', nr_employed['n'],',', y['n'] ) 
FROM blog_backup_data
WHERE customer_id['s'] > 0 ; 

After creating an external table, you need to read data. You then use the INSERT OVERWRITE DIRECTORY ~ SELECT command to write CSV data to the S3 path that you designated as the data source for Amazon SageMaker.

Depending on your requirements, you can eliminate or process the columns in the SELECT clause in this step to optimize data analysis. For example, you might remove some columns that have unpredictable correlations with the target value because keeping the wrong columns might expose your model to “overfitting” during the training. In this post, customer_id  columns is removed. Overfitting can make your prediction weak. More information about overfitting can be found in the topic Model Fit: Underfitting vs. Overfitting in the Amazon ML documentation.

Automation script: Renew the Amazon SageMaker model

After the CSV data is replaced and ready to use, create a new model artifact for Amazon SageMaker with the updated dataset on S3.  For renewing model artifact, you must create a new training job.  Training jobs can be run using the AWS SDK ( for example, Amazon SageMaker boto3 ) or the Amazon SageMaker Python SDK that can be installed with “pip install sagemaker” command as well as the AWS CLI for Amazon SageMaker described in this post.

In addition, consider how to smoothly renew your existing model without service impact, because your model is called by applications in real time. To do this, you need to create a new endpoint configuration first and update a current endpoint with the endpoint configuration that is just created.

#!/bin/bash
## Define variable 
REGION=$2
DTTIME=`date +%Y-%m-%d-%H-%M-%S`
ROLE="<your AmazonSageMaker-ExecutionRole>" 


# Select containers image based on region.  
case "$REGION" in
"us-west-2" )
    IMAGE="174872318107.dkr.ecr.us-west-2.amazonaws.com/linear-learner:latest"
    ;;
"us-east-1" )
    IMAGE="382416733822.dkr.ecr.us-east-1.amazonaws.com/linear-learner:latest" 
    ;;
"us-east-2" )
    IMAGE="404615174143.dkr.ecr.us-east-2.amazonaws.com/linear-learner:latest" 
    ;;
"eu-west-1" )
    IMAGE="438346466558.dkr.ecr.eu-west-1.amazonaws.com/linear-learner:latest" 
    ;;
 *)
    echo "Invalid Region Name"
    exit 1 ;  
esac

# Start training job and creating model artifact 
TRAINING_JOB_NAME=TRAIN-${DTTIME} 
S3OUTPUT="s3://<your bucket name>/model/" 
INSTANCETYPE="ml.m4.xlarge"
INSTANCECOUNT=1
VOLUMESIZE=5 
aws sagemaker create-training-job --training-job-name ${TRAINING_JOB_NAME} --region ${REGION}  --algorithm-specification TrainingImage=${IMAGE},TrainingInputMode=File --role-arn ${ROLE}  --input-data-config '[{ "ChannelName": "train", "DataSource": { "S3DataSource": { "S3DataType": "S3Prefix", "S3Uri": "s3://<your bucket name>/<datasource path>/", "S3DataDistributionType": "FullyReplicated" } }, "ContentType": "text/csv", "CompressionType": "None" , "RecordWrapperType": "None"  }]'  --output-data-config S3OutputPath=${S3OUTPUT} --resource-config  InstanceType=${INSTANCETYPE},InstanceCount=${INSTANCECOUNT},VolumeSizeInGB=${VOLUMESIZE} --stopping-condition MaxRuntimeInSeconds=120 --hyper-parameters feature_dim=20,predictor_type=binary_classifier  

# Wait until job completed 
aws sagemaker wait training-job-completed-or-stopped --training-job-name ${TRAINING_JOB_NAME}  --region ${REGION}

# Get newly created model artifact and create model
MODELARTIFACT=`aws sagemaker describe-training-job --training-job-name ${TRAINING_JOB_NAME} --region ${REGION}  --query 'ModelArtifacts.S3ModelArtifacts' --output text `
MODELNAME=MODEL-${DTTIME}
aws sagemaker create-model --region ${REGION} --model-name ${MODELNAME}  --primary-container Image=${IMAGE},ModelDataUrl=${MODELARTIFACT}  --execution-role-arn ${ROLE}

# create a new endpoint configuration 
CONFIGNAME=CONFIG-${DTTIME}
aws sagemaker  create-endpoint-config --region ${REGION} --endpoint-config-name ${CONFIGNAME}  --production-variants  VariantName=Users,ModelName=${MODELNAME},InitialInstanceCount=1,InstanceType=ml.m4.xlarge

# create or update the endpoint
STATUS=`aws sagemaker describe-endpoint --endpoint-name  ServiceEndpoint --query 'EndpointStatus' --output text --region ${REGION} `
if [[ $STATUS -ne "InService" ]] ;
then
    aws sagemaker  create-endpoint --endpoint-name  ServiceEndpoint  --endpoint-config-name ${CONFIGNAME} --region ${REGION}    
else
    aws sagemaker  update-endpoint --endpoint-name  ServiceEndpoint  --endpoint-config-name ${CONFIGNAME} --region ${REGION}
fi

Grant permission

Before you execute the script, you must grant proper permission to Data Pipeline. Data Pipeline uses the DataPipelineDefaultResourceRole role by default. I added the following policy to DataPipelineDefaultResourceRole to allow Data Pipeline to create, delete, and update the Amazon SageMaker model and data source in the script.

{
 "Version": "2012-10-17",
 "Statement": [
 {
 "Effect": "Allow",
 "Action": [
 "sagemaker:CreateTrainingJob",
 "sagemaker:DescribeTrainingJob",
 "sagemaker:CreateModel",
 "sagemaker:CreateEndpointConfig",
 "sagemaker:DescribeEndpoint",
 "sagemaker:CreateEndpoint",
 "sagemaker:UpdateEndpoint",
 "iam:PassRole"
 ],
 "Resource": "*"
 }
 ]
}

Use real-time prediction

After you deploy a model into production using Amazon SageMaker hosting services, your client applications use this API to get inferences from the model hosted at the specified endpoint. This approach is useful for interactive web, mobile, or desktop applications.

Following, I provide a simple Python code example that queries against Amazon SageMaker endpoint URL with its name (“ServiceEndpoint”) and then uses them for real-time prediction.

=== Python sample for real-time prediction ===

#!/usr/bin/env python
import boto3
import json 

client = boto3.client('sagemaker-runtime', region_name ='<your region>' )
new_customer_info = '34,10,2,4,1,2,1,1,6,3,190,1,3,4,3,-1.7,94.055,-39.8,0.715,4991.6'
response = client.invoke_endpoint(
    EndpointName='ServiceEndpoint',
    Body=new_customer_info, 
    ContentType='text/csv'
)
result = json.loads(response['Body'].read().decode())
print(result)
--- output(response) ---
{u'predictions': [{u'score': 0.7528127431869507, u'predicted_label': 1.0}]}

Solution summary

The solution takes the following steps:

  1. Data Pipeline exports DynamoDB table data into S3. The original JSON data should be kept to recover the table in the rare event that this is needed. Data Pipeline then converts JSON to CSV so that Amazon SageMaker can read the data.Note: You should select only meaningful attributes when you convert CSV. For example, if you judge that the “campaign” attribute is not correlated, you can eliminate this attribute from the CSV.
  2. Train the Amazon SageMaker model with the new data source.
  3. When a new customer comes to your site, you can judge how likely it is for this customer to subscribe to your new product based on “predictedScores” provided by Amazon SageMaker.
  4. If the new user subscribes your new product, your application must update the attribute “y” to the value 1 (for yes). This updated data is provided for the next model renewal as a new data source. It serves to improve the accuracy of your prediction. With each new entry, your application can become smarter and deliver better predictions.

Running ad hoc queries using Amazon Athena

Amazon Athena is a serverless query service that makes it easy to analyze large amounts of data stored in Amazon S3 using standard SQL. Athena is useful for examining data and collecting statistics or informative summaries about data. You can also use the powerful analytic functions of Presto, as described in the topic Aggregate Functions of Presto in the Presto documentation.

With the Data Pipeline scheduled activity, recent CSV data is always located in S3 so that you can run ad hoc queries against the data using Amazon Athena. I show this with example SQL statements following. For an in-depth description of this process, see the post Interactive SQL Queries for Data in Amazon S3 on the AWS News Blog. 

Creating an Amazon Athena table and running it

Simply, you can create an EXTERNAL table for the CSV data on S3 in Amazon Athena Management Console.

=== Table Creation ===
CREATE EXTERNAL TABLE datasource (
 age int, 
 job string, 
 marital string , 
 education string, 
 default string, 
 housing string, 
 loan string, 
 contact string, 
 month string, 
 day_of_week string, 
 duration int, 
 campaign int, 
 pdays int , 
 previous int , 
 poutcome string, 
 emp_var_rate double, 
 cons_price_idx double,
 cons_conf_idx double, 
 euribor3m double, 
 nr_employed double, 
 y int 
)
ROW FORMAT DELIMITED 
FIELDS TERMINATED BY ',' ESCAPED BY '\\' LINES TERMINATED BY '\n' 
LOCATION 's3://<your bucket name>/<datasource path>/';

The following query calculates the correlation coefficient between the target attribute and other attributes using Amazon Athena.

=== Sample Query ===

SELECT corr(age,y) AS correlation_age_and_target, 
 corr(duration,y) AS correlation_duration_and_target, 
 corr(campaign,y) AS correlation_campaign_and_target,
 corr(contact,y) AS correlation_contact_and_target
FROM ( SELECT age , duration , campaign , y , 
 CASE WHEN contact = 'telephone' THEN 1 ELSE 0 END AS contact 
 FROM datasource 
 ) datasource ;

Conclusion

In this post, I introduce an example of how to analyze data in DynamoDB by using table data in Amazon S3 to optimize DynamoDB table read capacity. You can then use the analyzed data as a new data source to train an Amazon SageMaker model for accurate real-time prediction. In addition, you can run ad hoc queries against the data on S3 using Amazon Athena. I also present how to automate these procedures by using Data Pipeline.

You can adapt this example to your specific use case at hand, and hopefully this post helps you accelerate your development. You can find more examples and use cases for Amazon SageMaker in the video AWS 2017: Introducing Amazon SageMaker on the AWS website.

 


Additional Reading

If you found this post useful, be sure to check out Serving Real-Time Machine Learning Predictions on Amazon EMR and Analyzing Data in S3 using Amazon Athena.

 


About the Author

Yong Seong Lee is a Cloud Support Engineer for AWS Big Data Services. He is interested in every technology related to data/databases and helping customers who have difficulties in using AWS services. His motto is “Enjoy life, be curious and have maximum experience.”

 

 

Easier way to control access to AWS regions using IAM policies

Post Syndicated from Sulay Shah original https://aws.amazon.com/blogs/security/easier-way-to-control-access-to-aws-regions-using-iam-policies/

We made it easier for you to comply with regulatory standards by controlling access to AWS Regions using IAM policies. For example, if your company requires users to create resources in a specific AWS region, you can now add a new condition to the IAM policies you attach to your IAM principal (user or role) to enforce this for all AWS services. In this post, I review conditions in policies, introduce the new condition, and review a policy example to demonstrate how you can control access across multiple AWS services to a specific region.

Condition concepts

Before I introduce the new condition, let’s review the condition element of an IAM policy. A condition is an optional IAM policy element that lets you specify special circumstances under which the policy grants or denies permission. A condition includes a condition key, operator, and value for the condition. There are two types of conditions: service-specific conditions and global conditions. Service-specific conditions are specific to certain actions in an AWS service. For example, the condition key ec2:InstanceType supports specific EC2 actions. Global conditions support all actions across all AWS services.

Now that I’ve reviewed the condition element in an IAM policy, let me introduce the new condition.

AWS:RequestedRegion condition key

The new global condition key, , supports all actions across all AWS services. You can use any string operator and specify any AWS region for its value.

Condition keyDescriptionOperator(s)Value
aws:RequestedRegionAllows you to specify the region to which the IAM principal (user or role) can make API callsAll string operators (for example, StringEqualsAny AWS region (for example, us-east-1)

I’ll now demonstrate the use of the new global condition key.

Example: Policy with region-level control

Let’s say a group of software developers in my organization is working on a project using Amazon EC2 and Amazon RDS. The project requires a web server running on an EC2 instance using Amazon Linux and a MySQL database instance in RDS. The developers also want to test Amazon Lambda, an event-driven platform, to retrieve data from the MySQL DB instance in RDS for future use.

My organization requires all the AWS resources to remain in the Frankfurt, eu-central-1, region. To make sure this project follows these guidelines, I create a single IAM policy for all the AWS services that this group is going to use and apply the new global condition key aws:RequestedRegion for all the services. This way I can ensure that any new EC2 instances launched or any database instances created using RDS are in Frankfurt. This policy also ensures that any Lambda functions this group creates for testing are also in the Frankfurt region.


{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": [
                "ec2:DescribeAccountAttributes",
                "ec2:DescribeAvailabilityZones",
                "ec2:DescribeInternetGateways",
                "ec2:DescribeSecurityGroups",
                "ec2:DescribeSubnets",
                "ec2:DescribeVpcAttribute",
                "ec2:DescribeVpcs",
                "ec2:DescribeInstances",
                "ec2:DescribeImages",
                "ec2:DescribeKeyPairs",
                "rds:Describe*",
                "iam:ListRolePolicies",
                "iam:ListRoles",
                "iam:GetRole",
                "iam:ListInstanceProfiles",
                "iam:AttachRolePolicy",
                "lambda:GetAccountSettings"
            ],
            "Resource": "*"
        },
        {
            "Effect": "Allow",
            "Action": [
                "ec2:RunInstances",
                "rds:CreateDBInstance",
                "rds:CreateDBCluster",
                "lambda:CreateFunction",
                "lambda:InvokeFunction"
            ],
            "Resource": "*",
      "Condition": {"StringEquals": {"aws:RequestedRegion": "eu-central-1"}}

        },
        {
            "Effect": "Allow",
            "Action": [
                "iam:PassRole"
            ],
            "Resource": "arn:aws:iam::account-id:role/*"
        }
    ]
}

The first statement in the above example contains all the read-only actions that let my developers use the console for EC2, RDS, and Lambda. The permissions for IAM-related actions are required to launch EC2 instances with a role, enable enhanced monitoring in RDS, and for AWS Lambda to assume the IAM execution role to execute the Lambda function. I’ve combined all the read-only actions into a single statement for simplicity. The second statement is where I give write access to my developers for the three services and restrict the write access to the Frankfurt region using the aws:RequestedRegion condition key. You can also list multiple AWS regions with the new condition key if your developers are allowed to create resources in multiple regions. The third statement grants permissions for the IAM action iam:PassRole required by AWS Lambda. For more information on allowing users to create a Lambda function, see Using Identity-Based Policies for AWS Lambda.

Summary

You can now use the aws:RequestedRegion global condition key in your IAM policies to specify the region to which the IAM principal (user or role) can invoke an API call. This capability makes it easier for you to restrict the AWS regions your IAM principals can use to comply with regulatory standards and improve account security. For more information about this global condition key and policy examples using aws:RequestedRegion, see the IAM documentation.

If you have comments about this post, submit them in the Comments section below. If you have questions about or suggestions for this solution, start a new thread on the IAM forum.

Want more AWS Security news? Follow us on Twitter.

No, Ray Ozzie hasn’t solved crypto backdoors

Post Syndicated from Robert Graham original https://blog.erratasec.com/2018/04/no-ray-ozzie-hasnt-solved-crypto.html

According to this Wired article, Ray Ozzie may have a solution to the crypto backdoor problem. No, he hasn’t. He’s only solving the part we already know how to solve. He’s deliberately ignoring the stuff we don’t know how to solve. We know how to make backdoors, we just don’t know how to secure them.

The vault doesn’t scale

Yes, Apple has a vault where they’ve successfully protected important keys. No, it doesn’t mean this vault scales. The more people and the more often you have to touch the vault, the less secure it becomes. We are talking thousands of requests per day from 100,000 different law enforcement agencies around the world. We are unlikely to protect this against incompetence and mistakes. We are definitely unable to secure this against deliberate attack.

A good analogy to Ozzie’s solution is LetsEncrypt for getting SSL certificates for your website, which is fairly scalable, using a private key locked in a vault for signing hundreds of thousands of certificates. That this scales seems to validate Ozzie’s proposal.

But at the same time, LetsEncrypt is easily subverted. LetsEncrypt uses DNS to verify your identity. But spoofing DNS is easy, as was recently shown in the recent BGP attack against a cryptocurrency. Attackers can create fraudulent SSL certificates with enough effort. We’ve got other protections against this, such as discovering and revoking the SSL bad certificate, so while damaging, it’s not catastrophic.

But with Ozzie’s scheme, equivalent attacks would be catastrophic, as it would lead to unlocking the phone and stealing all of somebody’s secrets.

In particular, consider what would happen if LetsEncrypt’s certificate was stolen (as Matthew Green points out). The consequence is that this would be detected and mass revocations would occur. If Ozzie’s master key were stolen, nothing would happen. Nobody would know, and evildoers would be able to freely decrypt phones. Ozzie claims his scheme can work because SSL works — but then his scheme includes none of the many protections necessary to make SSL work.

What I’m trying to show here is that in a lab, it all looks nice and pretty, but when attacked at scale, things break down — quickly. We have so much experience with failure at scale that we can judge Ozzie’s scheme as woefully incomplete. It’s not even up to the standard of SSL, and we have a long list of SSL problems.

Cryptography is about people more than math

We have a mathematically pure encryption algorithm called the “One Time Pad”. It can’t ever be broken, provably so with mathematics.

It’s also perfectly useless, as it’s not something humans can use. That’s why we use AES, which is vastly less secure (anything you encrypt today can probably be decrypted in 100 years). AES can be used by humans whereas One Time Pads cannot be. (I learned the fallacy of One Time Pad’s on my grandfather’s knee — he was a WW II codebreaker who broke German messages trying to futz with One Time Pads).

The same is true with Ozzie’s scheme. It focuses on the mathematical model but ignores the human element. We already know how to solve the mathematical problem in a hundred different ways. The part we don’t know how to secure is the human element.

How do we know the law enforcement person is who they say they are? How do we know the “trusted Apple employee” can’t be bribed? How can the law enforcement agent communicate securely with the Apple employee?

You think these things are theoretical, but they aren’t. Consider financial transactions. It used to be common that you could just email your bank/broker to wire funds into an account for such things as buying a house. Hackers have subverted that, intercepting messages, changing account numbers, and stealing millions. Most banks/brokers require additional verification before doing such transfers.

Let me repeat: Ozzie has only solved the part we already know how to solve. He hasn’t addressed these issues that confound us.

We still can’t secure security, much less secure backdoors

We already know how to decrypt iPhones: just wait a year or two for somebody to discover a vulnerability. FBI claims it’s “going dark”, but that’s only for timely decryption of phones. If they are willing to wait a year or two a vulnerability will eventually be found that allows decryption.

That’s what’s happened with the “GrayKey” device that’s been all over the news lately. Apple is fixing it so that it won’t work on new phones, but it works on old phones.

Ozzie’s solution is based on the assumption that iPhones are already secure against things like GrayKey. Like his assumption “if Apple already has a vault for private keys, then we have such vaults for backdoor keys”, Ozzie is saying “if Apple already had secure hardware/software to secure the phone, then we can use the same stuff to secure the backdoors”. But we don’t really have secure vaults and we don’t really have secure hardware/software to secure the phone.

Again, to stress this point, Ozzie is solving the part we already know how to solve, but ignoring the stuff we don’t know how to solve. His solution is insecure for the same reason phones are already insecure.

Locked phones aren’t the problem

Phones are general purpose computers. That means anybody can install an encryption app on the phone regardless of whatever other security the phone might provide. The police are powerless to stop this. Even if they make such encryption crime, then criminals will still use encryption.

That leads to a strange situation that the only data the FBI will be able to decrypt is that of people who believe they are innocent. Those who know they are guilty will install encryption apps like Signal that have no backdoors.

In the past this was rare, as people found learning new apps a barrier. These days, apps like Signal are so easy even drug dealers can figure out how to use them.

We know how to get Apple to give us a backdoor, just pass a law forcing them to. It may look like Ozzie’s scheme, it may be something more secure designed by Apple’s engineers. Sure, it will weaken security on the phone for everyone, but those who truly care will just install Signal. But again we are back to the problem that Ozzie’s solving the problem we know how to solve while ignoring the much larger problem, that of preventing people from installing their own encryption.

The FBI isn’t necessarily the problem

Ozzie phrases his solution in terms of U.S. law enforcement. Well, what about Europe? What about Russia? What about China? What about North Korea?

Technology is borderless. A solution in the United States that allows “legitimate” law enforcement requests will inevitably be used by repressive states for what we believe would be “illegitimate” law enforcement requests.

Ozzie sees himself as the hero helping law enforcement protect 300 million American citizens. He doesn’t see himself what he really is, the villain helping oppress 1.4 billion Chinese, 144 million Russians, and another couple billion living in oppressive governments around the world.

Conclusion

Ozzie pretends the problem is political, that he’s created a solution that appeases both sides. He hasn’t. He’s solved the problem we already know how to solve. He’s ignored all the problems we struggle with, the problems we claim make secure backdoors essentially impossible. I’ve listed some in this post, but there are many more. Any famous person can create a solution that convinces fawning editors at Wired Magazine, but if Ozzie wants to move forward he’s going to have to work harder to appease doubting cryptographers.

Tips for Success: GDPR Lessons Learned

Post Syndicated from Chad Woolf original https://aws.amazon.com/blogs/security/tips-for-success-gdpr-lessons-learned/

Security is our top priority at AWS, and from the beginning we have built security into the fabric of our services. With the introduction of GDPR (which becomes enforceable on May 25 of 2018), privacy and data protection have become even more ingrained into our security-centered culture. Three weeks ago, well ahead of the deadline, we announced that all AWS services are compliant with GDPR, meaning you can use AWS as a data processor as a way to help solve your GDPR challenges (be sure to visit our GDPR Center for additional information).

When it comes to GDPR compliance, many customers are progressing nicely and much of the initial trepidation is gone. In my interactions with customers on this topic, a few themes have emerged as universal:

  • GDPR is important. You need to have a plan in place if you process personal data of EU data subjects, not only because it’s good governance, but because GDPR does carry significant penalties for non-compliance.
  • Solving this can be complex, potentially involving a lot of personnel and multiple tools. Your GDPR process will also likely span across disciplines – impacting people, processes, and technology.
  • Each customer is unique, and there are many methodologies around assessing your compliance with GDPR. It’s important to be aware of your own individual business attributes.

I thought it might be helpful to share some of our own lessons learned. In our experience in solving the GDPR challenge, the following were keys to our success:

  1. Get your senior leadership involved. We have a regular cadence of detailed status conversations about GDPR with our CEO, Andy Jassy. GDPR is high stakes, and the AWS leadership team knows it. If GDPR doesn’t have the attention it needs with the visibility of top management today, it’s time to escalate.
  2. Centralize the GDPR efforts. Driving all work streams centrally is key. This may sound obvious, but managing this in a distributed manner may result in duplicative effort and/or team members moving in a different direction.
  3. The most important single partner in solving GDPR is your legal team. Having non-legal people make assumptions about how to interpret GDPR for your unique environment is both risky and a potential waste of time and resources. You want to avoid analysis paralysis by getting proper legal advice, collaborating on a direction, and then moving forward with the proper urgency.
  4. Collaborate closely with tech leadership. The “process” people in your organization, the ones who already know how to approach governance problems, are typically comfortable jumping right in to GDPR. But technical teams, including data owners, have set up their software for business application. They may not even know what kind of data they are storing, processing, or transferring to other parts of the business. In the GDPR exercise they need to be aware of (or at least help facilitate) the tracking of data and data elements between systems. This isn’t a typical ask for technical teams, so be prepared to educate and to fully understand data flow.
  5. Don’t live by the established checklists. There are multiple methodologies to solving the compliance challenges of GDPR. At AWS, we ended up establishing core requirements, mapped out by data controller and data processor functions and then, in partnership with legal, decided upon a group of projects based on our known current state. Be careful about using a set methodology, tool or questionnaire to govern your efforts. These generic assessments can help educate, but letting them drive or limit your work could lead to missing something that is key to your own compliance. In this sense, a generic, “one size fits all” solution might not be helpful.
  6. Don’t be afraid to challenge prior orthodoxy. Many times we changed course based on new information. You shouldn’t be afraid to scrap an effort if you determine it’s not working. You should also not be afraid to escalate issues to senior leadership when needed. This is an executive issue.
  7. Look for ways to leverage your work beyond this compliance activity. GDPR requires serious effort, but are the results limited to GDPR compliance? Certainly not. You can use GDPR workflows as a way to ensure better governance moving forward. Privacy and security will require work for the foreseeable future, so make your governance program scalable and usable for other purposes.

One last tip that has made all the difference: think about protecting data subjects and work backwards from there. Customer focus drives us to ask, “what would customers and data subjects want and expect us to do?” Taking GDPR from a pure legal or compliance standpoint may be technically sufficient, but we believe the objectives of security and personal data protection require a more comprehensive view, and you can most effectively shape that view by starting with the individuals GDPR was meant to protect.

If you would like to find out more about our experiences, as well as how we can help you in your efforts, please reach out to us today.

-Chad Woolf

Vice President, AWS Security Assurance

Interested in additional AWS Security news? Follow the AWS Security Blog on Twitter.

Computer Alarm that Triggers When Lid Is Opened

Post Syndicated from Bruce Schneier original https://www.schneier.com/blog/archives/2018/04/computer_alarm_.html

Do Not Disturb” is a Macintosh app that send an alert when the lid is opened. The idea is to detect computer tampering.

Wired article:

Do Not Disturb goes a step further than just the push notification. Using the Do Not Disturb iOS app, a notified user can send themselves a picture snapped with the laptop’s webcam to catch the perpetrator in the act, or they can shut down the computer remotely. The app can also be configured to take more custom actions like sending an email, recording screen activity, and keeping logs of commands executed on the machine.

Can someone please make one of these for Windows?

Implement continuous integration and delivery of serverless AWS Glue ETL applications using AWS Developer Tools

Post Syndicated from Prasad Alle original https://aws.amazon.com/blogs/big-data/implement-continuous-integration-and-delivery-of-serverless-aws-glue-etl-applications-using-aws-developer-tools/

AWS Glue is an increasingly popular way to develop serverless ETL (extract, transform, and load) applications for big data and data lake workloads. Organizations that transform their ETL applications to cloud-based, serverless ETL architectures need a seamless, end-to-end continuous integration and continuous delivery (CI/CD) pipeline: from source code, to build, to deployment, to product delivery. Having a good CI/CD pipeline can help your organization discover bugs before they reach production and deliver updates more frequently. It can also help developers write quality code and automate the ETL job release management process, mitigate risk, and more.

AWS Glue is a fully managed data catalog and ETL service. It simplifies and automates the difficult and time-consuming tasks of data discovery, conversion, and job scheduling. AWS Glue crawls your data sources and constructs a data catalog using pre-built classifiers for popular data formats and data types, including CSV, Apache Parquet, JSON, and more.

When you are developing ETL applications using AWS Glue, you might come across some of the following CI/CD challenges:

  • Iterative development with unit tests
  • Continuous integration and build
  • Pushing the ETL pipeline to a test environment
  • Pushing the ETL pipeline to a production environment
  • Testing ETL applications using real data (live test)
  • Exploring and validating data

In this post, I walk you through a solution that implements a CI/CD pipeline for serverless AWS Glue ETL applications supported by AWS Developer Tools (including AWS CodePipeline, AWS CodeCommit, and AWS CodeBuild) and AWS CloudFormation.

Solution overview

The following diagram shows the pipeline workflow:

This solution uses AWS CodePipeline, which lets you orchestrate and automate the test and deploy stages for ETL application source code. The solution consists of a pipeline that contains the following stages:

1.) Source Control: In this stage, the AWS Glue ETL job source code and the AWS CloudFormation template file for deploying the ETL jobs are both committed to version control. I chose to use AWS CodeCommit for version control.

To get the ETL job source code and AWS CloudFormation template, download the gluedemoetl.zip file. This solution is developed based on a previous post, Build a Data Lake Foundation with AWS Glue and Amazon S3.

2.) LiveTest: In this stage, all resources—including AWS Glue crawlers, jobs, S3 buckets, roles, and other resources that are required for the solution—are provisioned, deployed, live tested, and cleaned up.

The LiveTest stage includes the following actions:

  • Deploy: In this action, all the resources that are required for this solution (crawlers, jobs, buckets, roles, and so on) are provisioned and deployed using an AWS CloudFormation template.
  • AutomatedLiveTest: In this action, all the AWS Glue crawlers and jobs are executed and data exploration and validation tests are performed. These validation tests include, but are not limited to, record counts in both raw tables and transformed tables in the data lake and any other business validations. I used AWS CodeBuild for this action.
  • LiveTestApproval: This action is included for the cases in which a pipeline administrator approval is required to deploy/promote the ETL applications to the next stage. The pipeline pauses in this action until an administrator manually approves the release.
  • LiveTestCleanup: In this action, all the LiveTest stage resources, including test crawlers, jobs, roles, and so on, are deleted using the AWS CloudFormation template. This action helps minimize cost by ensuring that the test resources exist only for the duration of the AutomatedLiveTest and LiveTestApproval

3.) DeployToProduction: In this stage, all the resources are deployed using the AWS CloudFormation template to the production environment.

Try it out

This code pipeline takes approximately 20 minutes to complete the LiveTest test stage (up to the LiveTest approval stage, in which manual approval is required).

To get started with this solution, choose Launch Stack:

This creates the CI/CD pipeline with all of its stages, as described earlier. It performs an initial commit of the sample AWS Glue ETL job source code to trigger the first release change.

In the AWS CloudFormation console, choose Create. After the template finishes creating resources, you see the pipeline name on the stack Outputs tab.

After that, open the CodePipeline console and select the newly created pipeline. Initially, your pipeline’s CodeCommit stage shows that the source action failed.

Allow a few minutes for your new pipeline to detect the initial commit applied by the CloudFormation stack creation. As soon as the commit is detected, your pipeline starts. You will see the successful stage completion status as soon as the CodeCommit source stage runs.

In the CodeCommit console, choose Code in the navigation pane to view the solution files.

Next, you can watch how the pipeline goes through the LiveTest stage of the deploy and AutomatedLiveTest actions, until it finally reaches the LiveTestApproval action.

At this point, if you check the AWS CloudFormation console, you can see that a new template has been deployed as part of the LiveTest deploy action.

At this point, make sure that the AWS Glue crawlers and the AWS Glue job ran successfully. Also check whether the corresponding databases and external tables have been created in the AWS Glue Data Catalog. Then verify that the data is validated using Amazon Athena, as shown following.

Open the AWS Glue console, and choose Databases in the navigation pane. You will see the following databases in the Data Catalog:

Open the Amazon Athena console, and run the following queries. Verify that the record counts are matching.

SELECT count(*) FROM "nycitytaxi_gluedemocicdtest"."data";
SELECT count(*) FROM "nytaxiparquet_gluedemocicdtest"."datalake";

The following shows the raw data:

The following shows the transformed data:

The pipeline pauses the action until the release is approved. After validating the data, manually approve the revision on the LiveTestApproval action on the CodePipeline console.

Add comments as needed, and choose Approve.

The LiveTestApproval stage now appears as Approved on the console.

After the revision is approved, the pipeline proceeds to use the AWS CloudFormation template to destroy the resources that were deployed in the LiveTest deploy action. This helps reduce cost and ensures a clean test environment on every deployment.

Production deployment is the final stage. In this stage, all the resources—AWS Glue crawlers, AWS Glue jobs, Amazon S3 buckets, roles, and so on—are provisioned and deployed to the production environment using the AWS CloudFormation template.

After successfully running the whole pipeline, feel free to experiment with it by changing the source code stored on AWS CodeCommit. For example, if you modify the AWS Glue ETL job to generate an error, it should make the AutomatedLiveTest action fail. Or if you change the AWS CloudFormation template to make its creation fail, it should affect the LiveTest deploy action. The objective of the pipeline is to guarantee that all changes that are deployed to production are guaranteed to work as expected.

Conclusion

In this post, you learned how easy it is to implement CI/CD for serverless AWS Glue ETL solutions with AWS developer tools like AWS CodePipeline and AWS CodeBuild at scale. Implementing such solutions can help you accelerate ETL development and testing at your organization.

If you have questions or suggestions, please comment below.

 


Additional Reading

If you found this post useful, be sure to check out Implement Continuous Integration and Delivery of Apache Spark Applications using AWS and Build a Data Lake Foundation with AWS Glue and Amazon S3.

 


About the Authors

Prasad Alle is a Senior Big Data Consultant with AWS Professional Services. He spends his time leading and building scalable, reliable Big data, Machine learning, Artificial Intelligence and IoT solutions for AWS Enterprise and Strategic customers. His interests extend to various technologies such as Advanced Edge Computing, Machine learning at Edge. In his spare time, he enjoys spending time with his family.

 
Luis Caro is a Big Data Consultant for AWS Professional Services. He works with our customers to provide guidance and technical assistance on big data projects, helping them improving the value of their solutions when using AWS.

 

 

 

Implementing safe AWS Lambda deployments with AWS CodeDeploy

Post Syndicated from Chris Munns original https://aws.amazon.com/blogs/compute/implementing-safe-aws-lambda-deployments-with-aws-codedeploy/

This post courtesy of George Mao, AWS Senior Serverless Specialist – Solutions Architect

AWS Lambda and AWS CodeDeploy recently made it possible to automatically shift incoming traffic between two function versions based on a preconfigured rollout strategy. This new feature allows you to gradually shift traffic to the new function. If there are any issues with the new code, you can quickly rollback and control the impact to your application.

Previously, you had to manually move 100% of traffic from the old version to the new version. Now, you can have CodeDeploy automatically execute pre- or post-deployment tests and automate a gradual rollout strategy. Traffic shifting is built right into the AWS Serverless Application Model (SAM), making it easy to define and deploy your traffic shifting capabilities. SAM is an extension of AWS CloudFormation that provides a simplified way of defining serverless applications.

In this post, I show you how to use SAM, CloudFormation, and CodeDeploy to accomplish an automated rollout strategy for safe Lambda deployments.

Scenario

For this walkthrough, you write a Lambda application that returns a count of the S3 buckets that you own. You deploy it and use it in production. Later on, you receive requirements that tell you that you need to change your Lambda application to count only buckets that begin with the letter “a”.

Before you make the change, you need to be sure that your new Lambda application works as expected. If it does have issues, you want to minimize the number of impacted users and roll back easily. To accomplish this, you create a deployment process that publishes the new Lambda function, but does not send any traffic to it. You use CodeDeploy to execute a PreTraffic test to ensure that your new function works as expected. After the test succeeds, CodeDeploy automatically shifts traffic gradually to the new version of the Lambda function.

Your Lambda function is exposed as a REST service via an Amazon API Gateway deployment. This makes it easy to test and integrate.

Prerequisites

To execute the SAM and CloudFormation deployment, you must have the following IAM permissions:

  • cloudformation:*
  • lambda:*
  • codedeploy:*
  • iam:create*

You may use the AWS SAM Local CLI or the AWS CLI to package and deploy your Lambda application. If you choose to use SAM Local, be sure to install it onto your system. For more information, see AWS SAM Local Installation.

All of the code used in this post can be found in this GitHub repository: https://github.com/aws-samples/aws-safe-lambda-deployments.

Walkthrough

For this post, use SAM to define your resources because it comes with built-in CodeDeploy support for safe Lambda deployments.  The deployment is handled and automated by CloudFormation.

SAM allows you to define your Serverless applications in a simple and concise fashion, because it automatically creates all necessary resources behind the scenes. For example, if you do not define an execution role for a Lambda function, SAM automatically creates one. SAM also creates the CodeDeploy application necessary to drive the traffic shifting, as well as the IAM service role that CodeDeploy uses to execute all actions.

Create a SAM template

To get started, write your SAM template and call it template.yaml.

AWSTemplateFormatVersion : '2010-09-09'
Transform: AWS::Serverless-2016-10-31
Description: An example SAM template for Lambda Safe Deployments.

Resources:

  returnS3Buckets:
    Type: AWS::Serverless::Function
    Properties:
      Handler: returnS3Buckets.handler
      Runtime: nodejs6.10
      AutoPublishAlias: live
      Policies:
        - Version: "2012-10-17"
          Statement: 
          - Effect: "Allow"
            Action: 
              - "s3:ListAllMyBuckets"
            Resource: '*'
      DeploymentPreference:
          Type: Linear10PercentEvery1Minute
          Hooks:
            PreTraffic: !Ref preTrafficHook
      Events:
        Api:
          Type: Api
          Properties:
            Path: /test
            Method: get

  preTrafficHook:
    Type: AWS::Serverless::Function
    Properties:
      Handler: preTrafficHook.handler
      Policies:
        - Version: "2012-10-17"
          Statement: 
          - Effect: "Allow"
            Action: 
              - "codedeploy:PutLifecycleEventHookExecutionStatus"
            Resource:
              !Sub 'arn:aws:codedeploy:${AWS::Region}:${AWS::AccountId}:deploymentgroup:${ServerlessDeploymentApplication}/*'
        - Version: "2012-10-17"
          Statement: 
          - Effect: "Allow"
            Action: 
              - "lambda:InvokeFunction"
            Resource: !Ref returnS3Buckets.Version
      Runtime: nodejs6.10
      FunctionName: 'CodeDeployHook_preTrafficHook'
      DeploymentPreference:
        Enabled: false
      Timeout: 5
      Environment:
        Variables:
          NewVersion: !Ref returnS3Buckets.Version

This template creates two functions:

  • returnS3Buckets
  • preTrafficHook

The returnS3Buckets function is where your application logic lives. It’s a simple piece of code that uses the AWS SDK for JavaScript in Node.JS to call the Amazon S3 listBuckets API action and return the number of buckets.

'use strict';

var AWS = require('aws-sdk');
var s3 = new AWS.S3();

exports.handler = (event, context, callback) => {
	console.log("I am here! " + context.functionName  +  ":"  +  context.functionVersion);

	s3.listBuckets(function (err, data){
		if(err){
			console.log(err, err.stack);
			callback(null, {
				statusCode: 500,
				body: "Failed!"
			});
		}
		else{
			var allBuckets = data.Buckets;

			console.log("Total buckets: " + allBuckets.length);
			callback(null, {
				statusCode: 200,
				body: allBuckets.length
			});
		}
	});	
}

Review the key parts of the SAM template that defines returnS3Buckets:

  • The AutoPublishAlias attribute instructs SAM to automatically publish a new version of the Lambda function for each new deployment and link it to the live alias.
  • The Policies attribute specifies additional policy statements that SAM adds onto the automatically generated IAM role for this function. The first statement provides the function with permission to call listBuckets.
  • The DeploymentPreference attribute configures the type of rollout pattern to use. In this case, you are shifting traffic in a linear fashion, moving 10% of traffic every minute to the new version. For more information about supported patterns, see Serverless Application Model: Traffic Shifting Configurations.
  • The Hooks attribute specifies that you want to execute the preTrafficHook Lambda function before CodeDeploy automatically begins shifting traffic. This function should perform validation testing on the newly deployed Lambda version. This function invokes the new Lambda function and checks the results. If you’re satisfied with the tests, instruct CodeDeploy to proceed with the rollout via an API call to: codedeploy.putLifecycleEventHookExecutionStatus.
  • The Events attribute defines an API-based event source that can trigger this function. It accepts requests on the /test path using an HTTP GET method.
'use strict';

const AWS = require('aws-sdk');
const codedeploy = new AWS.CodeDeploy({apiVersion: '2014-10-06'});
var lambda = new AWS.Lambda();

exports.handler = (event, context, callback) => {

	console.log("Entering PreTraffic Hook!");
	
	// Read the DeploymentId & LifecycleEventHookExecutionId from the event payload
    var deploymentId = event.DeploymentId;
	var lifecycleEventHookExecutionId = event.LifecycleEventHookExecutionId;

	var functionToTest = process.env.NewVersion;
	console.log("Testing new function version: " + functionToTest);

	// Perform validation of the newly deployed Lambda version
	var lambdaParams = {
		FunctionName: functionToTest,
		InvocationType: "RequestResponse"
	};

	var lambdaResult = "Failed";
	lambda.invoke(lambdaParams, function(err, data) {
		if (err){	// an error occurred
			console.log(err, err.stack);
			lambdaResult = "Failed";
		}
		else{	// successful response
			var result = JSON.parse(data.Payload);
			console.log("Result: " +  JSON.stringify(result));

			// Check the response for valid results
			// The response will be a JSON payload with statusCode and body properties. ie:
			// {
			//		"statusCode": 200,
			//		"body": 51
			// }
			if(result.body == 9){	
				lambdaResult = "Succeeded";
				console.log ("Validation testing succeeded!");
			}
			else{
				lambdaResult = "Failed";
				console.log ("Validation testing failed!");
			}

			// Complete the PreTraffic Hook by sending CodeDeploy the validation status
			var params = {
				deploymentId: deploymentId,
				lifecycleEventHookExecutionId: lifecycleEventHookExecutionId,
				status: lambdaResult // status can be 'Succeeded' or 'Failed'
			};
			
			// Pass AWS CodeDeploy the prepared validation test results.
			codedeploy.putLifecycleEventHookExecutionStatus(params, function(err, data) {
				if (err) {
					// Validation failed.
					console.log('CodeDeploy Status update failed');
					console.log(err, err.stack);
					callback("CodeDeploy Status update failed");
				} else {
					// Validation succeeded.
					console.log('Codedeploy status updated successfully');
					callback(null, 'Codedeploy status updated successfully');
				}
			});
		}  
	});
}

The hook is hardcoded to check that the number of S3 buckets returned is 9.

Review the key parts of the SAM template that defines preTrafficHook:

  • The Policies attribute specifies additional policy statements that SAM adds onto the automatically generated IAM role for this function. The first statement provides permissions to call the CodeDeploy PutLifecycleEventHookExecutionStatus API action. The second statement provides permissions to invoke the specific version of the returnS3Buckets function to test
  • This function has traffic shifting features disabled by setting the DeploymentPreference option to false.
  • The FunctionName attribute explicitly tells CloudFormation what to name the function. Otherwise, CloudFormation creates the function with the default naming convention: [stackName]-[FunctionName]-[uniqueID].  Name the function with the “CodeDeployHook_” prefix because the CodeDeployServiceRole role only allows InvokeFunction on functions named with that prefix.
  • Set the Timeout attribute to allow enough time to complete your validation tests.
  • Use an environment variable to inject the ARN of the newest deployed version of the returnS3Buckets function. The ARN allows the function to know the specific version to invoke and perform validation testing on.

Deploy the function

Your SAM template is all set and the code is written—you’re ready to deploy the function for the first time. Here’s how to do it via the SAM CLI. Replace “sam” with “cloudformation” to use CloudFormation instead.

First, package the function. This command returns a CloudFormation importable file, packaged.yaml.

sam package –template-file template.yaml –s3-bucket mybucket –output-template-file packaged.yaml

Now deploy everything:

sam deploy –template-file packaged.yaml –stack-name mySafeDeployStack –capabilities CAPABILITY_IAM

At this point, both Lambda functions have been deployed within the CloudFormation stack mySafeDeployStack. The returnS3Buckets has been deployed as Version 1:

SAM automatically created a few things, including the CodeDeploy application, with the deployment pattern that you specified (Linear10PercentEvery1Minute). There is currently one deployment group, with no action, because no deployments have occurred. SAM also created the IAM service role that this CodeDeploy application uses:

There is a single managed policy attached to this role, which allows CodeDeploy to invoke any Lambda function that begins with “CodeDeployHook_”.

An API has been set up called safeDeployStack. It targets your Lambda function with the /test resource using the GET method. When you test the endpoint, API Gateway executes the returnS3Buckets function and it returns the number of S3 buckets that you own. In this case, it’s 51.

Publish a new Lambda function version

Now implement the requirements change, which is to make returnS3Buckets count only buckets that begin with the letter “a”. The code now looks like the following (see returnS3BucketsNew.js in GitHub):

'use strict';

var AWS = require('aws-sdk');
var s3 = new AWS.S3();

exports.handler = (event, context, callback) => {
	console.log("I am here! " + context.functionName  +  ":"  +  context.functionVersion);

	s3.listBuckets(function (err, data){
		if(err){
			console.log(err, err.stack);
			callback(null, {
				statusCode: 500,
				body: "Failed!"
			});
		}
		else{
			var allBuckets = data.Buckets;

			console.log("Total buckets: " + allBuckets.length);
			//callback(null, allBuckets.length);

			//  New Code begins here
			var counter=0;
			for(var i  in allBuckets){
				if(allBuckets[i].Name[0] === "a")
					counter++;
			}
			console.log("Total buckets starting with a: " + counter);

			callback(null, {
				statusCode: 200,
				body: counter
			});
			
		}
	});	
}

Repackage and redeploy with the same two commands as earlier:

sam package –template-file template.yaml –s3-bucket mybucket –output-template-file packaged.yaml
	
sam deploy –template-file packaged.yaml –stack-name mySafeDeployStack –capabilities CAPABILITY_IAM

CloudFormation understands that this is a stack update instead of an entirely new stack. You can see that reflected in the CloudFormation console:

During the update, CloudFormation deploys the new Lambda function as version 2 and adds it to the “live” alias. There is no traffic routing there yet. CodeDeploy now takes over to begin the safe deployment process.

The first thing CodeDeploy does is invoke the preTrafficHook function. Verify that this happened by reviewing the Lambda logs and metrics:

The function should progress successfully, invoke Version 2 of returnS3Buckets, and finally invoke the CodeDeploy API with a success code. After this occurs, CodeDeploy begins the predefined rollout strategy. Open the CodeDeploy console to review the deployment progress (Linear10PercentEvery1Minute):

Verify the traffic shift

During the deployment, verify that the traffic shift has started to occur by running the test periodically. As the deployment shifts towards the new version, a larger percentage of the responses return 9 instead of 51. These numbers match the S3 buckets.

A minute later, you see 10% more traffic shifting to the new version. The whole process takes 10 minutes to complete. After completion, open the Lambda console and verify that the “live” alias now points to version 2:

After 10 minutes, the deployment is complete and CodeDeploy signals success to CloudFormation and completes the stack update.

Check the results

If you invoke the function alias manually, you see the results of the new implementation.

aws lambda invoke –function [lambda arn to live alias] out.txt

You can also execute the prod stage of your API and verify the results by issuing an HTTP GET to the invoke URL:

Summary

This post has shown you how you can safely automate your Lambda deployments using the Lambda traffic shifting feature. You used the Serverless Application Model (SAM) to define your Lambda functions and configured CodeDeploy to manage your deployment patterns. Finally, you used CloudFormation to automate the deployment and updates to your function and PreTraffic hook.

Now that you know all about this new feature, you’re ready to begin automating Lambda deployments with confidence that things will work as designed. I look forward to hearing about what you’ve built with the AWS Serverless Platform.

Audit Trail Overview

Post Syndicated from Bozho original https://techblog.bozho.net/audit-trail-overview/

As part of my current project (secure audit trail) I decided to make a survey about the use of audit trail “in the wild”.

I haven’t written in details about this project of mine (unlike with some other projects). Mostly because it’s commercial and I don’t want to use my blog as a direct promotion channel (though I am doing that at the moment, ironically). But the aim of this post is to shed some light on how audit trail is used.

The survey can be found here. The questions are basically: does your current project have audit trail functionality, and if yes, is it protected from tampering. If not – do you think you should have such functionality.

The results are interesting (although with only around 50 respondents)

So more than half of the systems (on which respondents are working) don’t have audit trail. While audit trail is recommended by information security and related standards, it may not find place in the “busy schedule” of a software project, even though it’s fairly easy to provide a trivial implementation (e.g. I’ve written how to quickly setup one with Hibernate and Spring)

A trivial implementation might do in many cases but if the audit log is critical (e.g. access to sensitive data, performing financial operations etc.), then relying on a trivial implementation might not be enough. In other words – if the sysadmin can access the database and delete or modify the audit trail, then it doesn’t serve much purpose. Hence the next question – how is the audit trail protected from tampering:

And apparently, from the less than 50% of projects with audit trail, around 50% don’t have technical guarantees that the audit trail can’t be tampered with. My guess is it’s more, because people have different understanding of what technical measures are sufficient. E.g. someone may think that digitally signing your log files (or log records) is sufficient, but in fact it isn’t, as whole files (or records) can be deleted (or fully replaced) without a way to detect that. Timestamping can help (and a good audit trail solution should have that), but it doesn’t guarantee the order of events or prevent a malicious actor from deleting or inserting fake ones. And if timestamping is done on a log file level, then any not-yet-timestamped log file is vulnerable to manipulation.

I’ve written about event logs before and their two flavours – event sourcing and audit trail. An event log can effectively be considered audit trail, but you’d need additional security to avoid the problems mentioned above.

So, let’s see what would various levels of security and usefulness of audit logs look like. There are many papers on the topic (e.g. this and this), and they often go into the intricate details of how logging should be implemented. I’ll try to give an overview of the approaches:

  • Regular logs – rely on regular INFO log statements in the production logs to look for hints of what has happened. This may be okay, but is harder to look for evidence (as there is non-auditable data in those log files as well), and it’s not very secure – usually logs are collected (e.g. with graylog) and whoever has access to the log collector’s database (or search engine in the case of Graylog), can manipulate the data and not be caught
  • Designated audit trail – whether it’s stored in the database or in logs files. It has the proper business-event level granularity, but again doesn’t prevent or detect tampering. With lower risk systems that may is perfectly okay.
  • Timestamped logs – whether it’s log files or (harder to implement) database records. Timestamping is good, but if it’s not an external service, a malicious actor can get access to the local timestamping service and issue fake timestamps to either re-timestamp tampered files. Even if the timestamping is not compromised, whole entries can be deleted. The fact that they are missing can sometimes be deduced based on other factors (e.g. hour of rotation), but regularly verifying that is extra effort and may not always be feasible.
  • Hash chaining – each entry (or sequence of log files) could be chained (just as blockchain transactions) – the next one having the hash of the previous one. This is a good solution (whether it’s local, external or 3rd party), but it has the risk of someone modifying or deleting a record, getting your entire chain and re-hashing it. All the checks will pass, but the data will not be correct
  • Hash chaining with anchoring – the head of the chain (the hash of the last entry/block) could be “anchored” to an external service that is outside the capabilities of a malicious actor. Ideally, a public blockchain, alternatively – paper, a public service (twitter), email, etc. That way a malicious actor can’t just rehash the whole chain, because any check against the external service would fail.
  • WORM storage (write once, ready many). You could send your audit logs almost directly to WORM storage, where it’s impossible to replace data. However, that is not ideal, as WORM storage can be slow and expensive. For example AWS Glacier has rather big retrieval times and searching through recent data makes it impractical. It’s actually cheaper than S3, for example, and you can have expiration policies. But having to support your own WORM storage is expensive. It is a good idea to eventually send the logs to WORM storage, but “fresh” audit trail should probably not be “archived” so that it’s searchable and some actionable insight can be gained from it.
  • All-in-one – applying all of the above “just in case” may be unnecessary for every project out there, but that’s what I decided to do at LogSentinel. Business-event granularity with timestamping, hash chaining, anchoring, and eventually putting to WORM storage – I think that provides both security guarantees and flexibility.

I hope the overview is useful and the results from the survey shed some light on how this aspect of information security is underestimated.

The post Audit Trail Overview appeared first on Bozho's tech blog.

Achieving Major Stability and Performance Improvements in Yahoo Mail with a Novel Redux Architecture

Post Syndicated from mikesefanov original https://yahooeng.tumblr.com/post/173062946866

yahoodevelopers:

By Mohit Goenka, Gnanavel Shanmugam, and Lance Welsh

At Yahoo Mail, we’re constantly striving to upgrade our product experience. We do this not only by adding new features based on our members’ feedback, but also by providing the best technical solutions to power the most engaging experiences. As such, we’ve recently introduced a number of novel and unique revisions to the way in which we use Redux that have resulted in significant stability and performance improvements. Developers may find our methods useful in achieving similar results in their apps.

Improvements to product metrics

Last year Yahoo Mail implemented a brand new architecture using Redux. Since then, we have transformed the overall architecture to reduce latencies in various operations, reduce JavaScript exceptions, and better synchronized states. As a result, the product is much faster and more stable.

Stability improvements:

  • when checking for new emails – 20%
  • when reading emails – 30%
  • when sending emails – 20%

Performance improvements:

  • 10% improvement in page load performance
  • 40% improvement in frame rendering time

We have also reduced API calls by approximately 20%.

How we use Redux in Yahoo Mail

Redux architecture is reliant on one large store that represents the application state. In a Redux cycle, action creators dispatch actions to change the state of the store. React Components then respond to those state changes. We’ve made some modifications on top of this architecture that are atypical in the React-Redux community.

For instance, when fetching data over the network, the traditional methodology is to use Thunk middleware. Yahoo Mail fetches data over the network from our API. Thunks would create an unnecessary and undesirable dependency between the action creators and our API. If and when the API changes, the action creators must then also change. To keep these concerns separate we dispatch the action payload from the action creator to store them in the Redux state for later processing by “action syncers”. Action syncers use the payload information from the store to make requests to the API and process responses. In other words, the action syncers form an API layer by interacting with the store. An additional benefit to keeping the concerns separate is that the API layer can change as the backend changes, thereby preventing such changes from bubbling back up into the action creators and components. This also allowed us to optimize the API calls by batching, deduping, and processing the requests only when the network is available. We applied similar strategies for handling other side effects like route handling and instrumentation. Overall, action syncers helped us to reduce our API calls by ~20% and bring down API errors by 20-30%.

Another change to the normal Redux architecture was made to avoid unnecessary props. The React-Redux community has learned to avoid passing unnecessary props from high-level components through multiple layers down to lower-level components (prop drilling) for rendering. We have introduced action enhancers middleware to avoid passing additional unnecessary props that are purely used when dispatching actions. Action enhancers add data to the action payload so that data does not have to come from the component when dispatching the action. This avoids the component from having to receive that data through props and has improved frame rendering by ~40%. The use of action enhancers also avoids writing utility functions to add commonly-used data to each action from action creators.

image

In our new architecture, the store reducers accept the dispatched action via action enhancers to update the state. The store then updates the UI, completing the action cycle. Action syncers then initiate the call to the backend APIs to synchronize local changes.

Conclusion

Our novel use of Redux in Yahoo Mail has led to significant user-facing benefits through a more performant application. It has also reduced development cycles for new features due to its simplified architecture. We’re excited to share our work with the community and would love to hear from anyone interested in learning more.

Using AWS Lambda and Amazon Comprehend for sentiment analysis

Post Syndicated from Chris Munns original https://aws.amazon.com/blogs/compute/using-aws-lambda-and-amazon-comprehend-for-sentiment-analysis/

This post courtesy of Giedrius Praspaliauskas, AWS Solutions Architect

Even with best IVR systems, customers get frustrated. What if you knew that 10 callers in your Amazon Connect contact flow were likely to say “Agent!” in frustration in the next 30 seconds? Would you like to get to them before that happens? What if your bot was smart enough to admit, “I’m sorry this isn’t helping. Let me find someone for you.”?

In this post, I show you how to use AWS Lambda and Amazon Comprehend for sentiment analysis to make your Amazon Lex bots in Amazon Connect more sympathetic.

Setting up a Lambda function for sentiment analysis

There are multiple natural language and text processing frameworks or services available to use with Lambda, including but not limited to Amazon Comprehend, TextBlob, Pattern, and NLTK. Pick one based on the nature of your system:  the type of interaction, languages supported, and so on. For this post, I picked Amazon Comprehend, which uses natural language processing (NLP) to extract insights and relationships in text.

The walkthrough in this post is just an example. In a full-scale implementation, you would likely implement a more nuanced approach. For example, you could keep the overall sentiment score through the conversation and act only when it reaches a certain threshold. It is worth noting that this Lambda function is not called for missed utterances, so there may be a gap between what is being analyzed and what was actually said.

The Lambda function is straightforward. It analyses the input transcript field of the Amazon Lex event. Based on the overall sentiment value, it generates a response message with next step instructions. When the sentiment is neutral, positive, or mixed, the response leaves it to Amazon Lex to decide what the next steps should be. It adds to the response overall sentiment value as an additional session attribute, along with slots’ values received as an input.

When the overall sentiment is negative, the function returns the dialog action, pointing to an escalation intent (specified in the environment variable ESCALATION_INTENT_NAME) or returns the fulfillment closure action with a failure state when the intent is not specified. In addition to actions or intents, the function returns a message, or prompt, to be provided to the customer before taking the next step. Based on the returned action, Amazon Connect can select the appropriate next step in a contact flow.

For this walkthrough, you create a Lambda function using the AWS Management Console:

  1. Open the Lambda console.
  2. Choose Create Function.
  3. Choose Author from scratch (no blueprint).
  4. For Runtime, choose Python 3.6.
  5. For Role, choose Create a custom role. The custom execution role allows the function to detect sentiments, create a log group, stream log events, and store the log events.
  6. Enter the following values:
    • For Role Description, enter Lambda execution role permissions.
    • For IAM Role, choose Create an IAM role.
    • For Role Name, enter LexSentimentAnalysisLambdaRole.
    • For Policy, use the following policy:
{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": [
                "logs:CreateLogGroup",
                "logs:CreateLogStream",
                "logs:PutLogEvents"
            ],
            "Resource": "arn:aws:logs:*:*:*"
        },
        {
            "Action": [
                "comprehend:DetectDominantLanguage",
                "comprehend:DetectSentiment"
            ],
            "Effect": "Allow",
            "Resource": "*"
        }
    ]
}
    1. Choose Create function.
    2. Copy/paste the following code to the editor window
import os, boto3

ESCALATION_INTENT_MESSAGE="Seems that you are having troubles with our service. Would you like to be transferred to the associate?"
FULFILMENT_CLOSURE_MESSAGE="Seems that you are having troubles with our service. Let me transfer you to the associate."

escalation_intent_name = os.getenv('ESACALATION_INTENT_NAME', None)

client = boto3.client('comprehend')

def lambda_handler(event, context):
    sentiment=client.detect_sentiment(Text=event['inputTranscript'],LanguageCode='en')['Sentiment']
    if sentiment=='NEGATIVE':
        if escalation_intent_name:
            result = {
                "sessionAttributes": {
                    "sentiment": sentiment
                    },
                    "dialogAction": {
                        "type": "ConfirmIntent", 
                        "message": {
                            "contentType": "PlainText", 
                            "content": ESCALATION_INTENT_MESSAGE
                        }, 
                    "intentName": escalation_intent_name
                    }
            }
        else:
            result = {
                "sessionAttributes": {
                    "sentiment": sentiment
                },
                "dialogAction": {
                    "type": "Close",
                    "fulfillmentState": "Failed",
                    "message": {
                            "contentType": "PlainText",
                            "content": FULFILMENT_CLOSURE_MESSAGE
                    }
                }
            }

    else:
        result ={
            "sessionAttributes": {
                "sentiment": sentiment
            },
            "dialogAction": {
                "type": "Delegate",
                "slots" : event["currentIntent"]["slots"]
            }
        }
    return result
  1. Below the code editor specify the environment variable ESCALATION_INTENT_NAME with a value of Escalate.

  1. Click on Save in the top right of the console.

Now you can test your function.

  1. Click Test at the top of the console.
  2. Configure a new test event using the following test event JSON:
{
  "messageVersion": "1.0",
  "invocationSource": "DialogCodeHook",
  "userId": "1234567890",
  "sessionAttributes": {},
  "bot": {
    "name": "BookSomething",
    "alias": "None",
    "version": "$LATEST"
  },
  "outputDialogMode": "Text",
  "currentIntent": {
    "name": "BookSomething",
    "slots": {
      "slot1": "None",
      "slot2": "None"
    },
    "confirmationStatus": "None"
  },
  "inputTranscript": "I want something"
}
  1. Click Create
  2. Click Test on the console

This message should return a response from Lambda with a sentiment session attribute of NEUTRAL.

However, if you change the input to “This is garbage!”, Lambda changes the dialog action to the escalation intent specified in the environment variable ESCALATION_INTENT_NAME.

Setting up Amazon Lex

Now that you have your Lambda function running, it is time to create the Amazon Lex bot. Use the BookTrip sample bot and call it BookSomething. The IAM role is automatically created on your behalf. Indicate that this bot is not subject to the COPPA, and choose Create. A few minutes later, the bot is ready.

Make the following changes to the default configuration of the bot:

  1. Add an intent with no associated slots. Name it Escalate.
  2. Specify the Lambda function for initialization and validation in the existing two intents (“BookCar” and “BookHotel”), at the same time giving Amazon Lex permission to invoke it.
  3. Leave the other configuration settings as they are and save the intents.

You are ready to build and publish this bot. Set a new alias, BookSomethingWithSentimentAnalysis. When the build finishes, test it.

As you see, sentiment analysis works!

Setting up Amazon Connect

Next, provision an Amazon Connect instance.

After the instance is created, you need to integrate the Amazon Lex bot created in the previous step. For more information, see the Amazon Lex section in the Configuring Your Amazon Connect Instance topic.  You may also want to look at the excellent post by Randall Hunt, New – Amazon Connect and Amazon Lex Integration.

Create a new contact flow, “Sentiment analysis walkthrough”:

  1. Log in into the Amazon Connect instance.
  2. Choose Create contact flow, Create transfer to agent flow.
  3. Add a Get customer input block, open the icon in the top left corner, and specify your Amazon Lex bot and its intents.
  4. Select the Text to speech audio prompt type and enter text for Amazon Connect to play at the beginning of the dialog.
  5. Choose Amazon Lex, enter your Amazon Lex bot name and the alias.
  6. Specify the intents to be used as dialog branches that a customer can choose: BookHotel, BookTrip, or Escalate.
  7. Add two Play prompt blocks and connect them to the customer input block.
    • If booking hotel or car intent is returned from the bot flow, play the corresponding prompt (“OK, will book it for you”) and initiate booking (in this walkthrough, just hang up after the prompt).
    • However, if escalation intent is returned (caused by the sentiment analysis results in the bot), play the prompt (“OK, transferring to an agent”) and initiate the transfer.
  8. Save and publish the contact flow.

As a result, you have a contact flow with a single customer input step and a text-to-speech prompt that uses the Amazon Lex bot. You expect one of the three intents returned:

Edit the phone number to associate the contact flow that you just created. It is now ready for testing. Call the phone number and check how your contact flow works.

Cleanup

Don’t forget to delete all the resources created during this walkthrough to avoid incurring any more costs:

  • Amazon Connect instance
  • Amazon Lex bot
  • Lambda function
  • IAM role LexSentimentAnalysisLambdaRole

Summary

In this walkthrough, you implemented sentiment analysis with a Lambda function. The function can be integrated into Amazon Lex and, as a result, into Amazon Connect. This approach gives you the flexibility to analyze user input and then act. You may find the following potential use cases of this approach to be of interest:

  • Extend the Lambda function to identify “hot” topics in the user input even if the sentiment is not negative and take action proactively. For example, switch to an escalation intent if a user mentioned “where is my order,” which may signal potential frustration.
  • Use Amazon Connect Streams to provide agent sentiment analysis results along with call transfer. Enable service tailored towards particular customer needs and sentiments.
  • Route calls to agents based on both skill set and sentiment.
  • Prioritize calls based on sentiment using multiple Amazon Connect queues instead of transferring directly to an agent.
  • Monitor quality and flag for review contact flows that result in high overall negative sentiment.
  • Implement sentiment and AI/ML based call analysis, such as a real-time recommendation engine. For more details, see Machine Learning on AWS.

If you have questions or suggestions, please comment below.