README.md

etl-backup

Back up Kubecost's ETL data to your local filesystem. Execs into the kubecost pod, tars the contents of the ETL files, and kubectl cp's them down to your local machine.

Usage:

With your kubectl connected to the cluster whose data you want to back up:

git clone https://github.com/kubecost/kubecost-utilities.git

cd kubecost-utilities/etl-backup

./download-etl.sh <kubecost-namespace>

Run ETL from backed-up data

If this is the first time using backed up data, create the relevant directory:

sudo mkdir -p /var/configs/db/etl
sudo chmod -R 755 /var/configs/db/etl

Move the given backed-up data into the directory, so that bingen is the first directory nested under etl:

cd kc-etl-tmp
sudo mv bingen /var/configs/db/etl

Now you shold be able to make an ETL and query it:

// Create and run an ETL instance using the default ETL settings, with file backup
// enabled so that the backed-up ETL files are loaded.
conf := DefaultETLConfig()
conf.FileBackupEnabled = true
etl := RunETL(DefaultMockAllocationSource(), DefaultMockAssetSource(), nil, conf)

// Let the ETL run for a second
time.Sleep(time.Second)

// Let 'er rip
asr, err := etl.QueryAllocation(s, e, &kubecost.AllocationQueryOptions{
	AggregateBy: []string{"cluster", "namespace"},
})