i have documents modeling creation of ticket, such as:
{ "number": 12, "created_at": "2015-07-01t12:16:17z", "closed_at": null, "state": "open" } at point in future, second document models closing event:
{ "number": 12, "created_at": "2015-07-01t12:16:17z", "closed_at": "2015-07-08t8:12:42z", "state": "closed" } problem: want draw history of opened tickets. in example above, i'd ticket number 12 contribute count on whole 2015-07-01 2015-07-08 timespan. tried:
- bucketing
date_histogramseems able give number of tickets created or closed on given date bucket. - scripted metrics seem allow me change metric computation, not particular bucketing of document.
this first day playing elastic search , kibana might missing obvious. especially, cannot tell if buckets act partitions (hence if document can in single bucket), , hence if problem can solved creating additional documents each datapoint want appear on graph.
additional note: have control on feeding process , schema if storing additional data can help, i'd avoid doing if possible.
though thats not big deal , either mantain hashing on basis of dates, or keep
created_at
as grouping key documents made on day , can distinguish , query them want !!
Comments
Post a Comment