Value of RIPEstat routing-history full-peers-seeing
The RIPEstat Data API for the Routing History call includes a key output field "full-peers-seeing". It is described this way: The number of RIS full-feed peers that saw this route. Often times the value is not a whole number. I'm wondering why that may be the case and how I should interpret it. Hopefully this is as good a place as any to ask the question and get an answer archived for posterity. Thanks in advance for any insight. John
Hi John, This is due to the data aggregation that is applied when the time range is larger than a snapshot, e.g. 12 days by default. When you look at the Routing History widget this is shown as (more) human readable values, https://stat.ripe.net/widget/routing-history In the Data API output, the important value is "time_granularity", with possible values of 12d, 2d and 8h. The smaller the time interval ("starttime" to "endtime") for the lookup is, the less likely you get a fractional part (other than 0). Admittedly this is not too clear from the Data API documentation (https://stat.ripe.net/docs/data_api#routing-history) and will be something for the time when we revamp the documentation. Thanks for you question! Christian On 25/04/2021 23:18, John Kristoff wrote:
The RIPEstat Data API for the Routing History call includes a key output field "full-peers-seeing". It is described this way:
The number of RIS full-feed peers that saw this route.
Often times the value is not a whole number. I'm wondering why that may be the case and how I should interpret it. Hopefully this is as good a place as any to ask the question and get an answer archived for posterity. Thanks in advance for any insight.
John
-- Christian Teuschel | Human behind RIPEstat Follow me on Twitter or not - up to you: @christian_toysh
participants (2)
-
Christian Teuschel
-
John Kristoff