NetApp Interface Template for Cacti

Templates, scripts for templates, scripts and requests for templates.

Moderators: Moderators, Developers

Post Reply
Posts: 1
Joined: Tue Feb 05, 2008 12:20 pm
Location: Seattle, WA

NetApp Interface Template for Cacti

#1 Post by andreas » Tue Feb 05, 2008 12:29 pm

NetApp Interfaces are a serious pain, and even after the netapp engineers posted about how cool cacti is, they still are not able to come up with a workable template for their whacked out mib. I have worked through the mibs, and created some CDEF functions to add the high and low bits for gig-e interfaces ((high_bit*4294967296) + low). But the graphs do not look right, they jump up and down by 100MB nearly every poll. If anyone has any idea why this might be please post here. Import the cacti_host_template_netapp.xml into cacti, and put the netapp_interface.xml file in ~cacti/resources/snmp_queries/
(10.94 KiB) Downloaded 1106 times
graph_image.png (36.79 KiB) Viewed 8806 times

User avatar
Cacti Guru User
Posts: 2326
Joined: Fri Sep 19, 2003 8:36 am
Location: Sophia-Antipolis, France

#2 Post by fmangeant » Tue Feb 05, 2008 3:37 pm

Moving to the Templates forum.
[color=green]HOWTOs[/color] :
[list][*][url=]Install and configure the Net-SNMP agent for Unix[/url]
[*][url=]Install and configure the Net-SNMP agent for Windows[/url]
[*][url=]Graph multiple servers using an SNMP proxy[/url][/list]
[color=green]Templates[/color] :
[list][*][url=]Multiple CPU usage for Linux[/url]
[*][url=]Memory & swap usage for Unix[/url][/list][/size]

Posts: 3
Joined: Wed Jan 31, 2007 3:56 pm
Location: Austin, TX

Re: NetApp Interface Template for Cacti

#3 Post by sciolist » Fri Feb 27, 2009 3:12 pm


I've been looking at this problem and I think I've found part of an explanation, but haven't figured out how to make things work out correctly. From the rrdcreate manpage:

is for continuous incrementing counters like the ifInOctets
counter in a router. The COUNTER data source assumes that the
counter never decreases, except when a counter overflows. The
update function takes the overflow into account. The counter
is stored as a per-second rate. When the counter overflows,
RRDtool checks if the overflow happened at the 32bit or 64bit
border and acts accordingly by adding an appropriate value to
the result.

(emphasis mine). So what needs to happen is for the RRD file to *not* add that "appropriate value" to the low counters. I'm think that removing the Maximum Value setting from the low counters would do it, but that seems to be difficult without re-implementing the data template. Another possibility is to change the data type to "GAUGE" instead of "COUNTER". I'll keep working on it.


-=Tom Nail

Post Reply