0

i have an archive which contains the minimum duration times of messages going through an enterprise service bus.

During night-time or on weekends it is possible, that there are no messages, so the minimum duration time is NULL. Because of this, i write NaN into my rrd-db.

If i now want to consolidate an archive with the MINIMUM-Function over 6 hours, i get a problem if in this 6 hours there is a NaN-Value. Because then the whole 6-hours timeframe will be saved as NaN.

Is there any option to use a MINIMUM-Function, which ignores NaN-Values?

(Replacing NaN-Values with 0 would distort my graphs and is therefore not a good solution.)

Any help is very appreciated. Thank you in advance for participating!

4

1 回答 1

0

我自己找到了解决方案。我对文档的了解还不够。要忽略 NaN / Unknown 值,您必须使用 xff(xfiles 因子)。xff 为 0.99 时,NaN 值将被合并函数忽略。

http://oss.oetiker.ch/rrdtool/doc/rrdcreate.en.html

RRA:平均 | 最小 | 最大 | 最后:xff:步骤:行

xff xfiles 因子定义了合并区间的哪一部分可以由UNKNOWN数据组成,而合并值仍被认为是已知的。它以允许的UNKNOWN PDP 与间隔中的 PDP 数量之比给出。因此,它的范围从 0 到 1(不包括在内)。

于 2013-06-13T16:26:54.823 回答