In cloudyr/aws.cloudtrail: AWS CloudTrail Client Package

AWS CloudTrail Client Package

aws.cloudtrail is a simple client package for the Amazon Web Services (AWS) CloudTrail REST API, which can be used to monitor use of AWS web services API calls by logging API requests in an S3 bucket.

To use the package, you will need an AWS account and to enter your credentials into R. Your keypair can be generated on the IAM Management Console under the heading Access Keys. Note that you only have access to your secret key once. After it is generated, you need to save it in a secure location. New keypairs can be generated at any time if yours has been lost, stolen, or forgotten. The aws.iam package profiles tools for working with IAM, including creating roles, users, groups, and credentials programmatically; it is not needed to use IAM credentials.

A detailed description of how credentials can be specified is provided at: https://github.com/cloudyr/aws.signature/. The easiest way is to simply set environment variables on the command line prior to starting R or via an Renviron.site or .Renviron file, which are used to set environment variables in R during startup (see ? Startup). They can be also set within R:

Sys.setenv("AWS_ACCESS_KEY_ID" = "mykey",
           "AWS_SECRET_ACCESS_KEY" = "mysecretkey",
           "AWS_DEFAULT_REGION" = "us-east-1",
           "AWS_SESSION_TOKEN" = "mytoken")

Code Examples

A CloudTrail is a log of API calls made to AWS. The service is incredibly easy to use. It simply requires creating a trail that defines where (i.e., in what AWS S3 bucket) the CloudTrail log should be stored.

To use CloudTrail, start by creating an S3 bucket. The s3 bucket should exist (perhaps created using aws.s3) and have write permissions granted to CloudTrail. An example permission document is provided by cloudtrail_s3policy(). This can be done using the aws.s3 package:

library("aws.cloudtrail")
library("aws.s3")

# create bucket
mybucket <- "my1stcloudtrailbucket"
if (!bucket_exists(mybucket)) {
    stopifnot(put_bucket(mybucket))
}

# attach CloudTrail policy to bucket
ctpolicy <- cloudtrail_s3policy(mybucket, "my_aws_id")
stopifnot(put_bucket_policy(mybucket, policy = ctpolicy))

Note: CloudTrail appears to be very picky about bucket naming. The bucket name should be lowercase ASCII characters, without special characters.

trail <- create_trail(name = "NewTrail", bucket = mybucket)

# see trail in list of trails
"NewTrail" %in% unlist(lapply(get_trails(), `[[`, "Name"))

Once a trail is created, it can be updated (e.g., to move it to a different bucket, to activate event notifications using SNS, etc.) using update_trail().

# move trail to another bucket
otherbucket <- "myotherexamplebucket"
stopifnot(put_bucket(otherbucket))
ctpolicy <- cloudtrail_s3policy(otherbucket, "my_aws_id")
stopifnot(put_bucket_policy(otherbucket, policy = ctpolicy))

update_trail(name = "NewTrail", bucket = otherbucket)

# send SNS notifications when log updated
library("aws.sns")
top <- create_topic("MyCloudTrailTopic")
set_topic_attrs(top, list(Policy = cloudtrail_snspolicy(top)))
update_trail(name = "NewTrail", sns_topic = top)

# log global calls (e.g., IAM)
update_trail(name = "NewTrail", global = TRUE)

Once created and configured, it is easy to start logging requests using start_logging() and stop logging using stop_logging():

start_logging(trail)
trail_status(trail)$IsLogging # check logging status
stop_logging(trail)

The log is simply an S3 object, saved in the named bucket. We can check for logs using get_bucket() and retrieve one of the logs as a data frame using get_object():

(objects <- get_bucket(otherbucket))
mylog <- rawConnection(get_object(objects[[2]]))
jsonlite::fromJSON(mylog)

If you're done with a trail, you can delete it and it will no longer show up in your trail list:

delete_trail(trail)
get_trails()

Installation

This package is not yet on CRAN. To install the latest development version you can install from the cloudyr drat repository:

# latest stable version
install.packages("aws.cloudtrail", repos = c(cloudyr = "http://cloudyr.github.io/drat", getOption("repos")))

Or, to pull a potentially unstable version directly from GitHub:

```R if (!require("remotes")) { install.packages("remotes") } remotes::install_github("cloudyr/aws.cloudtrail")

---

cloudyr/aws.cloudtrail documentation built on Jan. 14, 2020, 4:54 a.m.