Hello all,
I've been recently twiddling about with some snmp monitoring of the funky netapps we have here. The big problem I've fallen over is the complete lack of details when it comes to lists of OID's and specific snmp queries you can make.
I need a list of all the things that can be queried on the Netapp, I've managed to pull bits and bobs from the mib and other sources, but the biggest one I've had so far is find entries for Bytes/s per interface (the same for collisions/errors). Are there such entries for load, tempuratures and so on? I've trawled now.netapp, a few FAQ's and also the archived mailing list without much joy :(
Any lists or pointers in the right direction would be appriciated.
Regards,
D.
FYI I pull the following from my Netcache, dont graph the filer atm but interested in ops per second if anyone knows where its hiding.. i pull my interface stats from my switches its easier! :)
Target[webcache-1-objects]: 1.3.6.1.4.1.789.1.8.3.1.0&.1.3.6.1.4.1.789.1.8.3.1.0:BLAH@1.2.3.4 Title[webcache-1-objects]: WebCache-1 C720 Objects Stored
Target[webcache-1-hitrate]: 1.3.6.1.4.1.789.1.8.3.14.2.0&1.3.6.1.4.1.789.1.8.3.14.1.0:BLAH@1.2.3.4 Title[webcache-1-hitrate]: Webcache-1 C720 Byte Hit Rate/Objects Hit Rate
Target[webcache-1-connects]: 1.3.6.1.4.1.789.1.8.3.3.0&1.3.6.1.4.1.789.1.8.3.4.0:BLAH@1.2.3.4 Title[webcache-1-connects]: WebCache-1 C720 Sockets (Servers/Clients)
Target[webcache-1-disk]: 1.3.6.1.4.1.789.1.5.4.1.4.1&1.3.6.1.4.1.789.1.5.4.1.3.1:BLAH@1.2.3.4 Title[webcache-1-disk]: WebCache-1 C720 Disk Usage (Used/Total)
-- Stephen J. Wilcox Internet Manager, Opal Telecom http://www.opaltelecom.co.uk/ Tel: 0161 222 2000 Fax: 0161 222 2008
On Mon, 11 Sep 2000, Demon Toasters Exploder wrote:
Hello all,
I've been recently twiddling about with some snmp monitoring of the funky netapps we have here. The big problem I've fallen over is the complete lack of details when it comes to lists of OID's and specific snmp queries you can make.
I need a list of all the things that can be queried on the Netapp, I've managed to pull bits and bobs from the mib and other sources, but the biggest one I've had so far is find entries for Bytes/s per interface (the same for collisions/errors). Are there such entries for load, tempuratures and so on? I've trawled now.netapp, a few FAQ's and also the archived mailing list without much joy :(
Any lists or pointers in the right direction would be appriciated.
Regards,
D.
-- Drav Sloan - Insane Senior Mail n News Admin - Platform Management Work Email: dsloan@demon.net / Fun Email: holborn@real-life.tm
Stephen J. Wilcox wrote:
FYI I pull the following from my Netcache, dont graph the filer atm but interested in ops per second if anyone knows where its hiding.. i pull my interface stats from my switches its easier! :)
Target[webcache-1-objects]: 1.3.6.1.4.1.789.1.8.3.1.0&.1.3.6.1.4.1.789.1.8.3.1.0:BLAH@1.2.3.4 Title[webcache-1-objects]: WebCache-1 C720 Objects Stored
Target[webcache-1-hitrate]: 1.3.6.1.4.1.789.1.8.3.14.2.0&1.3.6.1.4.1.789.1.8.3.14.1.0:BLAH@1.2.3.4 Title[webcache-1-hitrate]: Webcache-1 C720 Byte Hit Rate/Objects Hit Rate
Target[webcache-1-connects]: 1.3.6.1.4.1.789.1.8.3.3.0&1.3.6.1.4.1.789.1.8.3.4.0:BLAH@1.2.3.4 Title[webcache-1-connects]: WebCache-1 C720 Sockets (Servers/Clients)
Target[webcache-1-disk]: 1.3.6.1.4.1.789.1.5.4.1.4.1&1.3.6.1.4.1.789.1.5.4.1.3.1:BLAH@1.2.3.4 Title[webcache-1-disk]: WebCache-1 C720 Disk Usage (Used/Total)
Coolies, I'll throw these at the web team. I should of mentioned in me last post that I'm after filer stats. Is there meant to be a document that lists these? Or is there a tool to retrieve them?
Hoping for some answers (before I pull my techie hair out) :)
Regards,
D.
Hey,
Target[webcache-1-connects]: 1.3.6.1.4.1.789.1.8.3.3.0&1.3.6.1.4.1.789.1.8.3.4.0:BLAH@1.2.3.4 Title[webcache-1-connects]: WebCache-1 C720 Sockets (Servers/Clients)
Target[webcache-1-disk]: 1.3.6.1.4.1.789.1.5.4.1.4.1&1.3.6.1.4.1.789.1.5.4.1.3.1:BLAH@1.2.3.4 Title[webcache-1-disk]: WebCache-1 C720 Disk Usage (Used/Total)
Coolies, I'll throw these at the web team. I should of mentioned in me last post that I'm after filer stats. Is there meant to be a document that lists these? Or is there a tool to retrieve them?
Hoping for some answers (before I pull my techie hair out) :)
If you head towards the Download Software link on NOW!, turn left at Tools & Utilities, and download the 'MRTG for Filers' package the config files in there should give you something to go on.
:)
-- Regards,
Jay Tribick Senior Systems Engineer Carrier1 Voice: +44 207 531 3874 Mobile: +44 7801 526 638
Jay Tribick wrote:
Hey,
Haye dude :)
If you head towards the Download Software link on NOW!, turn left at Tools & Utilities, and download the 'MRTG for Filers' package the config files in there should give you something to go on.
Yeah I've downloaded that and used some of it (seems the actual OIDs are hidden somewhere deep withing mrtg/dat/conf files). Not been able to put my thumb on it, and I'd prefer to get a list of _all_ things to query. I'm actually setting up cricket and would love to have the complete list of things to log.
I've pulled some stuff off my filer for KB in/out and a few others the netapp MRTG does, but the big problemmo is data/second per interface instance and if such groovy things such as temperature can be measured.
Ta,
D.
Hey,
Haye dude :)
Duuude (:
If you head towards the Download Software link on NOW!, turn left at Tools & Utilities, and download the 'MRTG for Filers' package the config files in there should give you something to go on.
Yeah I've downloaded that and used some of it (seems the actual OIDs are hidden somewhere deep withing mrtg/dat/conf files). Not been able to put my thumb on it, and I'd prefer to get a list of _all_ things to query. I'm actually setting up cricket and would love to have the complete list of things to log.
Either read the MIB itself or..
I've pulled some stuff off my filer for KB in/out and a few others the netapp MRTG does, but the big problemmo is data/second per interface instance and if such groovy things such as temperature can be measured.
..do an 'snmpwalk host community' on the Filer and it'll spit out a load of interesting variables to play with - the ones you're looking for are:
interfaces.ifTable.ifEntry.ifInOctets.1 = 0 interfaces.ifTable.ifEntry.ifInOctets.2 = 0 interfaces.ifTable.ifEntry.ifInOctets.3 = 749712240 interfaces.ifTable.ifEntry.ifInOctets.4 = 32760 interfaces.ifTable.ifEntry.ifInOctets.5 = 229349808 interfaces.ifTable.ifEntry.ifInOctets.6 = 0 interfaces.ifTable.ifEntry.ifInOctets.7 = 610888 interfaces.ifTable.ifEntry.ifInOctets.8 = 0 interfaces.ifTable.ifEntry.ifInOctets.9 = 979079258 interfaces.ifTable.ifEntry.ifOutOctets.1 = 0 interfaces.ifTable.ifEntry.ifOutOctets.2 = 126 interfaces.ifTable.ifEntry.ifOutOctets.3 = 4203907540 interfaces.ifTable.ifEntry.ifOutOctets.4 = 126 interfaces.ifTable.ifEntry.ifOutOctets.5 = 3278641 interfaces.ifTable.ifEntry.ifOutOctets.6 = 0 interfaces.ifTable.ifEntry.ifOutOctets.7 = 610888 interfaces.ifTable.ifEntry.ifOutOctets.8 = 0 interfaces.ifTable.ifEntry.ifOutOctets.9 = 4207186433
-- Regards,
Jay Tribick Senior Systems Engineer Carrier1 Voice: +44 207 531 3874 Mobile: +44 7801 526 638
Jay Tribick wrote:
Either read the MIB itself or..
Ewww. Back to that nasty thing...
..do an 'snmpwalk host community' on the Filer and it'll spit out a load of interesting variables to play with - the ones you're looking for are:
Yup done that one. Now I thought I had figured by pulling these stats. But my take on it was that was the bytes per interface _since boot_ for that interface? Is that correct? I ask, because on our filers these values are insanly large, and if taken as a Octects/s then apparently our filers are shifting 25Pb/s - man wish they did :))
So, basically the choices are MiB or snmpwalk'ing. Okay here's one: how do I convert 'interfaces.ifTable.ifEntry.ifInOctets.1' to it's corresponding OID?
Cheers All,
D.
..do an 'snmpwalk host community' on the Filer and it'll spit out a load of interesting variables to play with - the ones you're looking for are:
Yup done that one. Now I thought I had figured by pulling these stats. But my take on it was that was the bytes per interface _since boot_ for that interface? Is that correct? I ask, because on our filers these values are insanly large, and if taken as a Octects/s then apparently our filers are shifting 25Pb/s - man wish they did :))
Yeh it's cumulative - it'll rollover when it gets to the maximum value for a 32-bit int, cricket should automatically take the roll-over into a account although you might have to do some tinkering with it to get it down to octets/sec.
So, basically the choices are MiB or snmpwalk'ing. Okay here's one: how do I convert 'interfaces.ifTable.ifEntry.ifInOctets.1' to it's corresponding OID?
Assuming you're using the ucd-snmp package, just type:
snmptranslate interfaces.ifTable.ifEntry.ifInOctets.1
:)
-- Regards,
Jay Tribick Senior Systems Engineer Carrier1 Voice: +44 207 531 3874 Mobile: +44 7801 526 638
Jay Tribick wrote:
values are insanly large, and if taken as a Octects/s then apparently our filers are shifting 25Pb/s - man wish they did :))
Yeh it's cumulative - it'll rollover when it gets to the maximum value for a 32-bit int, cricket should automatically take the roll-over into a account although you might have to do some tinkering with it to get it down to octets/sec.
Ah. Poo. Okay I'll go read the (*shudder*) cricket docs again and see if I can pull anything out of the hat. I hope it has the concept of converting cumulative to avg oct/s. Ho Hum.
Assuming you're using the ucd-snmp package, just type:
snmptranslate interfaces.ifTable.ifEntry.ifInOctets.1
Sweet, can't tell I'm an snmp newbie can ya? :)
D.
Yeh it's cumulative - it'll rollover when it gets to the maximum value for a 32-bit int, cricket should automatically take the roll-over into a account although you might have to do some tinkering with it to get it down to octets/sec.
Ah. Poo. Okay I'll go read the (*shudder*) cricket docs again and see if I can pull anything out of the hat. I hope it has the concept of converting cumulative to avg oct/s. Ho Hum.
Should work fine - let me know if you run into problems and I'll forward you my cricket configs.
-- Regards,
Jay Tribick Senior Systems Engineer Carrier1 Voice: +44 207 531 3874 Mobile: +44 7801 526 638
* Demon Toasters Exploder (toasters@demon.net) done spit this rhetoric:
Hello all,
I've been recently twiddling about with some snmp monitoring of the funky netapps we have here. The big problem I've fallen over is the complete lack of details when it comes to lists of OID's and specific snmp queries you can make.
I need a list of all the things that can be queried on the Netapp, I've managed to pull bits and bobs from the mib and other sources, but the biggest one I've had so far is find entries for Bytes/s per interface (the same for collisions/errors). Are there such entries for load, tempuratures and so on? I've trawled now.netapp, a few FAQ's and also the archived mailing list without much joy :(
Any lists or pointers in the right direction would be appriciated.
There's actually a list generated with the creation of every release, but they forgot to put it in /etc/mib with the mib (this is fixed 6.0.) The file is called traps.dat, and you can get it by sending mail to support@netapp.com to request it for the version that you are running.
-
Demon Toasters Exploder -
Drav Sloan -
Jay Tribick -
Jerry Talkington -
Stephen J. Wilcox