[rrd-users] Re: Unexpected NaN values

jeffrey.j.petter at verizon.com jeffrey.j.petter at verizon.com
Thu Nov 2 23:24:27 MET 2006


Alex wrote...."Are you doing everything the same?  No different schedule,
no
difference in processing data?  Then the problem has to be in
the rrd file.

Run "rrdtool info" on one of the working files, and on the one
not working.  Compare.  Look especially at minimum and maximum
values, as well as heartbeat.

You are _sure_ you do get data from your device, right?"

You are correct, everything I am doing and have done, is the same for all
files. They were all created the same way, with the same script, they are
all being updated with the same script.

I ran "rrdtool info" against my problem file and one of my good ones, and
I'll be the first to admit that I really don't understand all that is in
there. However, there was quite a large number of NaN's in the problem one.
Here is a quick paste comparing similar parts in each file:
good:
< ds[accepted].last_ds = "195653"
< ds[accepted].value = 1.6802259630e+01
< ds[accepted].unknown_sec = 0
bad:
> ds[accepted].last_ds = "292670430"
> ds[accepted].value = NaN
> ds[accepted].unknown_sec = 38

good:
< ds[dropped].last_ds = "14865"
< ds[dropped].value = 1.3230125693e-01
< ds[dropped].unknown_sec = 0
bad
> ds[dropped].last_ds = "467805739"
> ds[dropped].value = NaN
> ds[dropped].unknown_sec = 38

The rra's were all NaN's with the maximum number of unknown_datapoints.

It would seem to me that if there is a value for last_ds, there should be a
value, but then that could mean the current value is NaN and the last was
valid. But, most likely that is not the case since all the data in the file
is NaN's.

In my script that performs the updates to rrd, I am writing the device
name, update date, time, values to a log file, so I know that data updates
were attempted. Perhaps I'll try my luck "rrdtool dump"ing from a good file
and restore it into a new one, and try to collect the data that I am not
able to now.

Thanks,
Jeff

--
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://lists.ee.ethz.ch/rrd-users
WebAdmin    http://lists.ee.ethz.ch/lsg2.cgi



More information about the rrd-users mailing list