This is intended to be run as a cron job or something as such for sanity checks on your server.

• disk usage check
• tcp connection check
• http connection check

export GOPATH=$HOME/go
export PATH=$PATH:$GOPATH/bin
export GOROOT=/usr/local/go      # assuming go is installed at /usr/local/go
export PATH=$PATH:$GOROOT/bin

If you haven't gotten godep yet, do that using

go get github.com/tools/godep

go get github.com/prasincs/alertif
cd $GOPATH/src/github.com/prasincs/alertif
make all

You can run the command using

./alertif -p <pagerduty service key> --disk -t 80 -i "/dev,/tmp" -s example,tcp,8888,dead

This means

• -p Run with a pagerduty service key that you've obtained from the site.
• --disk enable checking disk usage
• -t,--disk-threshold If there's any disk that's using more than 80 percent disk usage, send PagerDuty Alert
• -i,--disk-ignore Ignores "/dev" and "/tmp" mountpoints
• -s,--service If the service is dead, sends alert. Note the syntax, it goes as follows: Name,Type,Port,Action
• -h, --hostname The Hostname you want to use for service checks.

A Service Command is a comma separated list. It is broken down into the following components:

"Name,Type,Port,Action"

The resulting alert would be sent under the name you've given the service. For http service type, the Action is the URL you want to GET.

• [DONE] Adding service monitoring
• Add end to end tests