Anyone interested in NetApp 8.2 templates - CDOT and 7-mode?

[vgrig]

CURRENT_DATA_SOURCE,ALL_DATA_SOURCES_NODUPS,/,100,* is given as example in the docs for "disk used" and "disk available" and NetApp only gives "disk used" and "disk total" (as qrV264KBytesUsed and qrV264KBytesLimit), so you're still getting wrong data: (used) / ((used + total) * 100 istead of correct one - used / total * 100...
i'm playing with CDEF a bit and i think the correct one is CURRENT_DATA_SOURCE,ALL_DATA_SOURCES_NODUPS,CURRENT_DATA_SOURCE,-,/,100,* - but nor sure. Try creating that CDEF as "_Current DS as Percentage of total" - see if the works better.

[aftershocks]

This is the version that doesn't work for me when I try and query.. 8.2.2P2 Cluster-Mode..

However, 8.3P1 works... Strange.. But when I do an snmpwalk on ".1.3.6.1.4.1.789.1.5.4.1", volumes come up! It just can't pick it up in Cacti.. Weird..

[vgrig]

I just noticed - if i enable user quotas on a qtree, all qtrees stop graphing. don't know why.

[aftershocks]

Humm.. I tried your CDEF (CURRENT_DATA_SOURCE,ALL_DATA_SOURCES_NODUPS,CURRENT_DATA_SOURCE,-,/,100,* ) but not getting correct data... Its kind of off by a lot... Maybe I'm doing something wrong here...

[vgrig]

Yeah, i'm having no luck either. I'm gonna step back for a while and come back to the problem later.

[vgrig]

OK, i think i got it this time.
First - there are 3 CDEFs involved, so create them first (i included 2 grap templates - one with and one without dependencies):
1. "_b as percentage of a" (cdef=b,a,/,100,*)
2. "_Make Current Data Source 0" (cdef=CURRENT_DATA_SOURCE,0,*)
3. "_Make Current Data Source 100" (cdef=CURRENT_DATA_SOURCE,CURRENT_DATA_SOURCE,-,100,+)

Then import "NetApp - QTree % 64-bit CDOT" graph template and assiciate it with "NetApp - QTrees 64-bit CDOT" data query.
After that (important) click on "NetApp - QTree % 64-bit CDOT" under "Associated Graph Templates" and make sure "Associated Data Templates" looks like below (somehow it picked up wrong data sources before - that's why we were getting wrong graph values).

% graphs track KB used graphs corectly now - as you can see below.

Hope this helps.

[aftershocks]

Thanks vgrig for your help!

[stormonts]

For me, everythign is working after importing the templates and putting the .xml files in the correct location, however my "SNMP - Interface Statistics" aren't returning stats on any of the LIFs. When I do a verbose query, I get:

+ Running data query [1].
+ Found type = '3' [SNMP Query].
+ Found data query XML file at '/usr/local/cacti-0.8.8h/resource/snmp_queries/interface.xml'
+ XML file parsed ok.
+ Executing SNMP get for num of indexes @ '.1.3.6.1.2.1.2.1.0' Index Count:
+ Executing SNMP walk for list of indexes @ '.1.3.6.1.2.1.2.2.1.1' Index Count: 0
+ No SNMP data returned

[vgrig]

@stormonts - that "SNMP - Interface Statistics" is not netapp specific: it's the same one that pulls switch interface stats. I guess netapp changed what OIDs return interface stats in CDOT, so netapp specific xml is needed.
honestly though - i always monitor switch ports for interface stats: much easier. so, i'm not sure i'll have time to spend to fix this.

one thing i'm trying to fix is qtree usage stops graphing if user quotas are present - they return eithe * or empty string and that breaks all qtree graphs.

[aftershocks]

Hi vgrig,

I have this working on version 8.3P1... However I have upgraded another NetApp to 8.3.2P4 and it stops working..

Do you think it's because NetApp changed some OIDs?

[vgrig]

What's not working? All qtree graphs? What about volumes graphs?

[aftershocks]

All of them, but I'm just interested in the volumes...

It's like they changed something in the versions and it's not able to query anything...

Do you by any chance know how i can somehow remap to the new OIDs? Or is it easier said than done?

[aftershocks]

So weird... When I do a snmpwalk on the v8.3.2P4 (not working in cacti) on .1.3.6.1.4.1.789.1.5.4.1 (volumes) - it queries and returns all the information... Same when I do it for v8.3P1... But somehow, it's not returning anything when I use your template on v8.3.2P4..

I thought NetApp changed their OIDS but nope.. The work.. But for some reason, the template doesn't pick it up...

All the OIDs from the netapp-volumes.xml file can be queried on the one that is not working in Cacti...

[aftershocks]

This is what I mean:

I run a verbose query and this is what it returns..

+ Running data query [27].
+ Found type = '3' [SNMP Query].
+ Found data query XML file at '/usr/share/cacti/site/resource/snmp_queries/netapp-volumes.xml'
+ XML file parsed ok.
+ <oid_num_indexes> missing in XML file, 'Index Count Changed' emulated by counting oid_index entries
+ Executing SNMP walk for list of indexes @ '.1.3.6.1.4.1.789.1.5.4.1' Index Count: 0
+ No SNMP data returned

Index Count: 0..

But when I run a snmpwalk on the same node.......


iso.3.6.1.4.1.789.1.5.4.1.1.1026 = INTEGER: 1026
iso.3.6.1.4.1.789.1.5.4.1.1.1027 = INTEGER: 1027
iso.3.6.1.4.1.789.1.5.4.1.1.1028 = INTEGER: 1028
iso.3.6.1.4.1.789.1.5.4.1.1.1029 = INTEGER: 1029
iso.3.6.1.4.1.789.1.5.4.1.1.1030 = INTEGER: 1030
iso.3.6.1.4.1.789.1.5.4.1.1.1031 = INTEGER: 1031
iso.3.6.1.4.1.789.1.5.4.1.1.1032 = INTEGER: 1032
iso.3.6.1.4.1.789.1.5.4.1.1.1033 = INTEGER: 1033
iso.3.6.1.4.1.789.1.5.4.1.1.1034 = INTEGER: 1034
iso.3.6.1.4.1.789.1.5.4.1.1.1035 = INTEGER: 1035
iso.3.6.1.4.1.789.1.5.4.1.1.1036 = INTEGER: 1036
iso.3.6.1.4.1.789.1.5.4.1.1.1037 = INTEGER: 1037
iso.3.6.1.4.1.789.1.5.4.1.1.1038 = INTEGER: 1038
iso.3.6.1.4.1.789.1.5.4.1.1.1039 = INTEGER: 1039
iso.3.6.1.4.1.789.1.5.4.1.1.1042 = INTEGER: 1042
iso.3.6.1.4.1.789.1.5.4.1.1.1043 = INTEGER: 1043
iso.3.6.1.4.1.789.1.5.4.1.1.1044 = INTEGER: 1044
iso.3.6.1.4.1.789.1.5.4.1.1.1045 = INTEGER: 1045
iso.3.6.1.4.1.789.1.5.4.1.1.1046 = INTEGER: 1046
iso.3.6.1.4.1.789.1.5.4.1.1.1047 = INTEGER: 1047
iso.3.6.1.4.1.789.1.5.4.1.1.1048 = INTEGER: 1048
iso.3.6.1.4.1.789.1.5.4.1.1.1049 = INTEGER: 1049
iso.3.6.1.4.1.789.1.5.4.1.1.1050 = INTEGER: 1050
iso.3.6.1.4.1.789.1.5.4.1.1.1051 = INTEGER: 1051
iso.3.6.1.4.1.789.1.5.4.1.1.1052 = INTEGER: 1052
iso.3.6.1.4.1.789.1.5.4.1.1.1053 = INTEGER: 1053
iso.3.6.1.4.1.789.1.5.4.1.1.1054 = INTEGER: 1054
iso.3.6.1.4.1.789.1.5.4.1.1.1055 = INTEGER: 1055
iso.3.6.1.4.1.789.1.5.4.1.1.1056 = INTEGER: 1056
iso.3.6.1.4.1.789.1.5.4.1.1.1057 = INTEGER: 1057
iso.3.6.1.4.1.789.1.5.4.1.1.1058 = INTEGER: 1058
iso.3.6.1.4.1.789.1.5.4.1.1.1059 = INTEGER: 1059
iso.3.6.1.4.1.789.1.5.4.1.1.1060 = INTEGER: 1060
iso.3.6.1.4.1.789.1.5.4.1.1.1061 = INTEGER: 1061
iso.3.6.1.4.1.789.1.5.4.1.1.1062 = INTEGER: 1062
iso.3.6.1.4.1.789.1.5.4.1.1.1063 = INTEGER: 1063
iso.3.6.1.4.1.789.1.5.4.1.1.1064 = INTEGER: 1064
iso.3.6.1.4.1.789.1.5.4.1.1.1065 = INTEGER: 1065
iso.3.6.1.4.1.789.1.5.4.1.1.1066 = INTEGER: 1066
iso.3.6.1.4.1.789.1.5.4.1.1.1067 = INTEGER: 1067
iso.3.6.1.4.1.789.1.5.4.1.1.1068 = INTEGER: 1068
iso.3.6.1.4.1.789.1.5.4.1.1.1069 = INTEGER: 1069
iso.3.6.1.4.1.789.1.5.4.1.1.1070 = INTEGER: 1070
iso.3.6.1.4.1.789.1.5.4.1.1.1071 = INTEGER: 1071
iso.3.6.1.4.1.789.1.5.4.1.1.1072 = INTEGER: 1072
iso.3.6.1.4.1.789.1.5.4.1.1.1073 = INTEGER: 1073
....
...
..

It sees everything... So I'm kinda lost...

[vgrig]

Strange. Minor version upgrade like that shouldn't really break things. You can try to compare NetApp MIB files from 2 versions.