NetApp 2.4TB volume

[slay]

I am using cacti-0.8.6i on Centos 4.4

I am graphing a NetApp FAS270 and everything is working OK except for /vol/vol1

This volume is 2.4TB and has 300GB free. The graph shows the 300GB free but shows Freeas 0.0 and Used as 0.0

If anyone else has seen this I would appreciate some pointers. I have tried adjusting the graph templates with the multiplication by 1024 and stuff but nothing I have tried gets rrdtool or cacti to show the TB limit.

There is only a small amount of data on the rtight side because I have recreated the host a bunch of times messing with it.

[rony]

SNMPv1 or SNMPv2?

[slay]

SNMPv1

I tried to get the netapp to respond to v2 but it wouldn't. Plus this is in production and I didn't want to change too much......

The files is running NetApp Release 6.5.2R1

--

[rony]

What is the block size on that volume?

[slay]

4096

[rony]

Um... The number of blocks should fit in a 32-bit integer and should work with SNMPv1.

Call Netapp?

[slay]

So maybe I am misunderstanding you.

What do you mean by block size? You mean the block size of the disks in the RAID group for vol1?

--

[rony]

Well, in SNMP world, disk space is reported as the number of blocks.

So, 2.4 TB is about 5.8 Million 4096 byte blocks.

The limit of SNMPv1 is 32-bit integer, which is 4294967296, so, the resulting 5.8 million is with in that bit range.

So, I guess the next step is posting a walk of the storage section.

Please get a snmpwalk of oid .1.2.6.1.2.1.25.2.3 and post it.

[slay]

Hey Rony

First I want to say thanks for taking your time to help me with this. I do appreciate it.

So I tried a snmpwalk with that OID against the filer like this:

Code: Select all

snmpwalk -v 1 -c <community> <hostname> .1.2.6.1.2.1.25.2.3

I got no reply. Basically back to a prompt with no output.

I did do a walk of .1 and grabbed it to a file and saw this at the beginning:

Code: Select all

SNMPv2-MIB::sysDescr.0 = STRING: NetApp Release 6.5.2R1: Wed Sep  8 15:58:37 PDT 2004
SNMPv2-MIB::sysObjectID.0 = OID: SNMPv2-SMI::enterprises.789.2.1
SNMPv2-MIB::sysUpTime.0 = Timeticks: (139861020) 16 days, 4:30:10.20
SNMPv2-MIB::sysContact.0 = STRING: no@no.com
SNMPv2-MIB::sysName.0 = STRING: hostname.domain.com
SNMPv2-MIB::sysLocation.0 = STRING: Server Room
SNMPv2-MIB::sysServices.0 = INTEGER: 72

Is the SNMPv2-MIB mean the NetApp is using v2c? When I try v2c with snmpwalk it times out:

Code: Select all

Timeout: No Response from hostname

So I am not sure at this point. I also see this info when I grep the walk for vol1:

Code: Select all

SNMPv2-SMI::enterprises.789.1.4.4.1.2.1 = STRING: "vol1"

[rony]

I want to see all oid's matching:

Code: Select all

SNMPv2-SMI::enterprises.789.1.4.4.1.2

[slay]

Took this from the .1 walk:

Code: Select all

[root@host tmp]# cat foo | grep SNMPv2-SMI::enterprises.789.1.4.4.1.2
SNMPv2-SMI::enterprises.789.1.4.4.1.2.1 = STRING: "vol1"
SNMPv2-SMI::enterprises.789.1.4.4.1.2.2 = STRING: "vol2"
SNMPv2-SMI::enterprises.789.1.4.4.1.2.3 = STRING: "vol0"

--

[pbulteel]

Something that might be related... I don't know and I might open another thread on this one.

I have a server with 6.6 T of disk space -- of which currently only 236 Gigs are being used.

My disk usage graphs show 219G Used and "NaN" available.

Have we somehow reached a limit?!

-P

[rony]

Something that might be related... I don't know and I might open another thread on this one.

I have a server with 6.6 T of disk space -- of which currently only 236 Gigs are being used.

My disk usage graphs show 219G Used and "NaN" available.

Have we somehow reached a limit?!

-P

The SNMPv1 limit, maybe... Or a SNMP agent limit..

[pbulteel]

looks like I have...

snmpwalk -v 1 -c public xxxxx .1.3.6.1.4.1.2021.9.1.7
UCD-SNMP-MIB::dskAvail.1 = INTEGER: 7425440
UCD-SNMP-MIB::dskAvail.2 = INTEGER: 1880764
UCD-SNMP-MIB::dskAvail.3 = INTEGER: -2147483648
snmpwalk -v 1 -c public xxxxx .1.3.6.1.4.1.2021.9.1.6
UCD-SNMP-MIB::dskTotal.1 = INTEGER: 10080520
UCD-SNMP-MIB::dskTotal.2 = INTEGER: 2016016
UCD-SNMP-MIB::dskTotal.3 = INTEGER: -2147483648
snmpwalk -v 1 -c public xxxxx .1.3.6.1.4.1.2021.9.1.8
UCD-SNMP-MIB::dskUsed.1 = INTEGER: 2143012
UCD-SNMP-MIB::dskUsed.2 = INTEGER: 32840
UCD-SNMP-MIB::dskUsed.3 = INTEGER: 230135092

-P

[rony]

Yep, looks like an overflow...

Will it respond to SNMPv2?