[rrd-users] Re: How to eliminata extrimely high values

Katsifos Nikolaos bakira at bakira.org
Mon Sep 30 12:53:18 MEST 2002


Dear Alex ....
First i have to thank you for your responce and for your time spending for
my problem.

>> First have a look at your "extremely high" values.  Perhaps you
>> have a configuration problem and should get rid of those values
>> altogether.
>> For instance: If you didn't set a proper MAX setting, you can expect
>> random high values when the device reboots or when the counters are
>> reset.  These values can be *very* high, such as terabytes per second.
>> If you suffer from this, don't try to work around the problem but
>> solve the problem.

I am not using RRDTOOL for routers/switches/SNMP-enabled-devices
monitoring..
I am using RRDTOOL to plot time values as GAUGE.
e.g
I request a DNS entry and then i plot the responce time.

It could take from 3 millisecs to 4000 millisecs etc etc.
That means there is no MAX for my values ;-(

BTW ... Thank you for your script advice...

> However:  How do you know that the 95th percentile doesn't result
> in another high value?

Actualy I  dont.. But my expirence said that.. i saw the plots and with this
percentile
i can drop all this high values from all of my plots

Another algorithm i am thinkig is the folowing
LIMIT=(AVERAGE+60)
Is it possible ???
Is it correct the folowing command ??
CDEF:V=value,0,AVERAGE,150,+,LIMIT

And a last question/wish to Tobi...
I have read the mailing list and i saw there is a need for persentile
support..
While it is imposible to do that with CDEF/RPN and i have to write a script
to calculate this
is it possible to add this function native in a next version of RRDTOOL???

Nikos Katsifos
p.s. Sorry again  for my bad english...

--
Unsubscribe mailto:rrd-users-request at list.ee.ethz.ch?subject=unsubscribe
Help        mailto:rrd-users-request at list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/rrd-users
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi



More information about the rrd-users mailing list