Graphite summarise() 函数结果对于不同的“from”值不一致

Question

I am using Graphite to record user login information.

When I run the following :

render?target=summarize(stats_counts.login.success,"1day")&format=json&from=-1days

I am getting the result :

[
  {
    "target": "summarize(stats_counts.login.success, \"1day\", \"sum\")",
    "datapoints": [
      [
        5,
        1435708800
      ],
      [
        21,
        1435795200
      ]
    ]
  }
]

But for the following query :

render?target=summarize(stats_counts.login.success,"1day")&format=json&from=-7days

I am getting the result :

[
  {
    "target": "summarize(stats_counts.login.success, \"1day\", \"sum\")",
    "datapoints": [
      [
        0,
        1435190400
      ],
      [
        1,
        1435276800
      ],
      [
        0,
        1435363200
      ],
      [
        0,
        1435449600
      ],
      [
        5,
        1435536000
      ],
      [
        16,
        1435622400
      ],
      [
        6,
        1435708800
      ],
      [
        21,
        1435795200
      ]
    ]
  }
]

Notice the value for the bucket : 1435708800 in both the results. In one result it is : 5 and in the second result it is : 6

In the first query I am trying to get the number of user logins per day over the last week and in the second one I am trying to get the number of user logins per day yesterday and today.

What is the reason for this difference ?

UPDATE

Graphite Version : 0.9.10 Retention Settings :

[carbon]
pattern = ^carbon\.
retentions = 60:90d

[real_time]
priority = 200
pattern = ^stats.*
retentions = 1:34560000

[stats]
priority = 110
pattern = .*
retentions = 1s:24h,1m:7d,10m:1y

Answer 1

Try adding allign to true, since based on time it varies the number of data points it get on the bucket interval.

By default, buckets are calculated by rounding to the nearest interval. This works well for intervals smaller than a day. For example, 22:32 will end up in the bucket 22:00-23:00 when the interval=1hour.

Passing alignToFrom=true will instead create buckets starting at the from time. In this case, the bucket for 22:32 depends on the from time. If from=6:30 then the 1hour bucket for 22:32 is 22:30-23:30.

"summarize(ex.cpe.ex.xxx,'30s','avg', true)"