Using IBM Connect:Direct® for Unix with AWS S3
This section contains all the information you need in order to use IBM Connect:Direct® for UNIX to configure processes for either uploading or downloading user files with Amazon S3.
- AWS S3 buckets
- Working with Traces
Using AWS S3 buckets
The copy step of the Connect:Direct Process statement supports uploading and downloading user files using the AWS S3 Object Store. The Connect:Direct Unix node must be installed on an EC2 instance in the same AWS region as the S3 bucket. The node installed on EC2 can be used as Pnode or Snode and either can read and write user files using S3.
- Changes in Process Language:
The Exits are specified in the FILE keyword of the Connect:Direct process language. The s3://myBucket represents the S3 bucket to store sample.txt in.
sample PROCESS SNODE=cd-ec2 SNODEID=(cd-proxy-user) upld COPY FROM ( FILE=sample.bin ) CKPT=10M TO ( FILE=s3://myBucket/sample.bin DISP=RPL ) PEND;
The process displayed above will upload a file named "sample.bin" from an on-premise Pnode to an Snode in the AWS cloud named cd-ec2. The file is stored in myBucket on S3. The AWS credentials configured for the cd-proxy-user are used to access S3
S3 checkpoint intervals are supported from 10M to 1G. If the value is out of range, a warning is logged to statistics.
User would need to consider bucket versioning on S3. When uploading a file to S3 the option to modify file (MOD) is not supported. Only NEW and RPL options can be set by the user.
In case of NEW – If Bucket versioning is on, then a new version of the file will be created on S3.
In case of RPL – If Bucket versioning is on, then latest version of the file will be replaced. If the file does not exist then it'll be created.
Working with Traces
Enabling traces for Connect:Direct Unix running on AWS is similar to how it’s done in an on-premise version. S3 traces are included when SMGR tracing is enabled