[rrd-developers] Posible bug/unexpected behaviour on templated update

[javier at rediris.es]

Hello,

a couple of weeks ago, I noticed some problems when using rrdupdate with
a template. Instead of list datasources in a different order, I pretended
to update only a subset of them. When doing this I noticed some problems,
and more systematic trials showed me that it only occurrs for counter and
derive types, not for gauges.
I also noticed some errors trying when using PREV during graphs creation,
but I didn't attempt any systematic efforts.
I solved this problem splitting the initial rrd, with the only drawback
that could happen that I need to cross two rrd for a graph (unavailable
feature to my known)

Last week, I tried to modify rrdupdate to met our needs and I discovered
the problem. In whichever way you call rrdupdate, all datasources are
set to unknown, and later the command line values are feeded, rewritting
the previous unknown values. As every datasource last_ds is rewritten,
those not included in the template gets an unknonw last_ds, leading
to uknown counter/derive values.

I'm not sure if my description is clear, but the problem is simple.
The solution is not so simple, because if we only update the last_ds
of the datasources in template, we arrive to a problem with the interval
since last update, producing unuseful (although not unknown) values
for counter/derive.

I've made a couple of trials and I think I'm able to fix the problem,
but I want be sure that anybody else is attempting the same and I also
prefer to get some 'quorum' before. Any comment or advice is very
wellcomed, but I'll assume that silence mean yes.

							Javi


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