For When You Can't Have The Real Thing
[ start | index | login ]
start > netapp > cacti counters

cacti counters

Created by dave. Last edited by dave, 9 years and 197 days ago. Viewed 12,497 times. #18
[diff] [history] [edit] [rdf]
labels

Cacti Counters of NetApp

>>This page is what I am working from.

Notes

I am working with cacti 8.7g, installed from RPMforge, on CentOS 5.5, at the moment (October 2010).

What the page tells you to do:

  • Import the host-template
  • copy the netapp-volumes.xml file to "the scripts directory". What this means is
# cd /tmp
# wget >>http://...netapp-volumes.xml
# dos2unix < netapp-volumes.xml > ~cacti/resource/snmp_queries/netapp-volumes.xml

(I don't know why the dos2unix voodoo works. But if you don't do it, you get errors saying Error parsing XML file into an array or, if you look in Console -> Data Queries -> NetApp Volumes, it will say Could not locate XML file. The file is there (proven by changing the macro to be the absolute path, then copy-pasting that file specifier to the shell and 'ls' on that) but doesn't get accepted until you dos2unix it.)

Now you must solve the following problems:

Problem: No Volumes or Aggregates are listed

To fix: go to the Create Graphs page, then run the volume query in debug mode. Then go back and look at the Create Graphs page, the volumes are listed.

Problem: No interfaces listed to create graphs for

To fix: go to the Create Graphs page, then run the interface query in debug mode. Then go back and look at the Create Graphs page, the interfaces are listed.

Problem: blank graphs for CPU, Cache Age, NFS Ops and CIFS Ops

Data is being populated in the .rrd file.

If you run the graph in debug mode, at the bottom it says: ERROR:the RRD does not contain an RRA matching the chosen CF

  • go Console -> Management -> Data Sources -> RRAs; for each name, edit it and make sure that AVERAGE, MAX, and LAST are selected (you will have to add LAST); then re-save. You may have to remove the affected RRD files from ~cacti/rra to get the additional information generated.
(For the purposes of Google linkage: >>Discussion -- halfway down the second page a post from "Gandalf" details what has to be done.)

My Status

Everything appears to work at this point. Well except for the part that aggregate and volume values larger than 2^31 (around 2.1TB) are being read as negative numbers, but that's possibly a weakness in snmp v1.

no comments | post comment
This is a collection of techical information, much of it learned the hard way. Consider it a lab book or a /info directory. I doubt much of it will be of use to anyone else.

Useful:


snipsnap.org | Copyright 2000-2002 Matthias L. Jugel and Stephan J. Schmidt