[rrd-users] Re: Weird peaks on loaded systems over NFS

[Andre Bonhote]

Good day again!

On Tue, Nov 12, 2002 at 08:56:36AM +0100, Andre Bonhote wrote:

[... snipped large explanation part ...]
 
> Ok. Can someone tell me what could be done to avoid this, or has anyone
> made a similar experience with rrdtool over NFS? For me, it's just plain
> incomprehensible that locking issues result in a peak. Are there some
> fine-tuning tipps to produce cleaner graphs, without giving a MAX?

Ok, I think I have to add a few things at this place. I've read some
articles in this list and rrd-devel about running rrdtool with the
.rrd's on a remote NFS server. I've read about the issues statd/lockd
and they've been solved from the beginning. What I did NOT find is the
reason for the peaks.

The scripts are reporting "Could not lock RRD". Well, that's one thing,
maybe related with NFS. For me it would be obvious that RRD would not
write into this file then. But what it does (at least what the graph
shows afterwards) is a peak of at least 5 times the average. Why this? I
mean, a gap inside the graph is one thing, we could live with that
maybe, but data inserted where no data should be is another thing.

Any guesses/hints/solutions?

TIA

André (still hoping for an answer)


-- 
Real programmers use "cat > a.out"

--
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