You will need to modify the source cblogger.go. This repo is only a scaffold. To be able to versionize and push your changes you should fork the repo and put it under your prefered version control.

The easiest would be to just clone the repo to your server.

git clone <repo url>

Of course you can also integrate it with your prefered deployment tools.

The repo contains an example init script that works on AWS AMI images. If you need to change the options with which the logger starts, you can edit it here.

The supported parameters are:

--logdir Where to put the csv files. The default is ./.

--logfile The name of the current log file. The default is csv.

--time The suffix each file gets. The option accepts the golang time strings. The default is 2006-01-02_15.

--location The time location. The default is UTC.

--s3 The s3 bucket name. By default this is unset, which means no rotation to s3. If you don't setup s3 rotation old files are not deleted. If you use s3 rotation only the last 12 files are kept locally. If rotation to s3 fails, the files gets renamed to failed_<filename> and gets excluded from deletion.

--interval How often to move the files out. The default is 3600 seconds. If you change the rotation interval you will have to change the --time parameter as well to avoid overwriting your files.

--port The port on which the cblogger runs. The default is 3000.

If you use an AWS AMI image the installation should be as simple as

sudo bash bootstrap.sh

This will do the following steps:

• update your system
• install the golang package
• install and configure nginx, add nginx to default runlevel and start nginx
• install the deamonize software
• build the binary, configure the init script, add to default runlevel and the start the software

If you want to use the hourly push to s3 you also need to configure the aws cli tool to use your credentials.

sudo su -
aws configure
<follow the dialog>

Please be aware that this setup will run this software as root. If you are not comfortable with it, you need to edit the init script and permissions.

Also if you want to use another webserver than nginx, you will need to perform the steps described in bootstrap.sh and replace the nginx steps with appropriate configuration steps for your prefered webserver.

If you want to add other placeholders to your csv files, you can do that by editing the source code. There is an array called paramList which holds all params you are interested in.

A little example: We set a callback in our adjust dashboard that reads:

http://your-domain/?app_name=my_app&event_name={event}

adjust would replace {event} with an appropriate value, resulting in a callback that looks something like:

http://your-domain/?app_name=my_app&event_name=f0ob4r

Now say you want your CSV to look like this:

"event_name","app_name"

So your paramList needs to look like that:

paramList = []string{
    "app_name", "event_name",
}
// trailing comma is important!

After you edited the source code, you can rebuild the binary and restart the software with

sudo bash build.sh