Make sure your storage-aggregator.py is running in order databases to compress down into 2nd and 3rd retention levels.

$ ls
league_number.wsp  user_number.wsp

Find out info on your database “league_number”
Seeing something like xFilesFactor .5 means that 50% of that time retention period has to show
if you have 10 second intervals but are only inputting once every 60 seconds then your xFilesFactor is only .2
Set the xFilesFactor = 0 if you don’t want to drop any of your data

$ /usr/local/bin/whisper-info.py league_number.wsp
maxRetention: 315360000
xFilesFactor: 0.5
aggregationMethod: average
fileSize: 63097960

Archive 0
retention: 21600
secondsPerPoint: 10
points: 2160
size: 25920
offset: 40

Archive 1
retention: 315360000
secondsPerPoint: 60
points: 5256000
size: 63072000
offset: 25960

Resize your whisper database if data is not being aggregated properly

$ /usr/local/bin/whisper-resize.py league_number.wsp
Usage: whisper-resize.py path timePerPoint:timeToStore [timePerPoint:timeToStore]*

timePerPoint and timeToStore specify lengths of time, for example:

60:1440      60 seconds per datapoint, 1440 datapoints = 1 day of retention
15m:8        15 minutes per datapoint, 8 datapoints = 2 hours of retention
1h:7d        1 hour per datapoint, 7 days of retention
12h:2y       12 hours per datapoint, 2 years of retention

If you have fucked up graphs that are truncating data on the second retention size
align them like so

$ /usr/local/bin/whisper-resize.py user_number.wsp --xFilesFactor=0 --aggregationMethod=average 10:6h 1min:10y

Make sure you set your storage-aggregation correctly and make sure it’s started

$ cp storage-aggregation.conf.example storage-aggregation.conf

$ /opt/graphite/bin/carbon-aggregator.py status
carbon-aggregator (instance a) is not running
$ /opt/graphite/bin/carbon-aggregator.py start
Starting carbon-aggregator (instance a)