• December 10, 2018, 06:07:52 PM
Welcome, Guest. Please login or register. Registration is free.
Did you miss your activation email?

Author Topic: PRTG CPU Monitoring  (Read 672 times)

0 Members and 1 Guest are viewing this topic.

Offline Abdul

  • Jr. Member
  • **
  • Posts: 35
PRTG CPU Monitoring
« on: July 07, 2017, 11:42:05 AM »
Hi Guys,

I am trying to monitor my 5500 and 4850 switch and stack CPU's using PRTG SNMP but I am struggling with the OID and Sensors. Has anyone done this before and can share the sensor type  and OID's used.

Thanks
Abdul


Offline TankII

  • Hero Member
  • *****
  • Posts: 544
Re: PRTG CPU Monitoring
« Reply #1 on: July 10, 2017, 10:23:14 AM »
CPUs 1-8 in a stack:
1.3.6.1.4.1.45.1.6.3.8.1.1.5 Instance 3.10.0
1.3.6.1.4.1.45.1.6.3.8.1.1.5 instance 3.20.0
... Continues until
1.3.6.1.4.1.45.1.6.3.8.1.1.5 instance 3.80.0

Memory:
1.3.6.1.4.1.45.1.6.3.8.1.1.9 instance 3.10.0
Repeat until you get to instance 3.80.0

CPU Temperature:
1.3.6.1.4.1.45.1.6.3.10 Instance 5.10.0
Then 5.20.0 through 5.80.0
Note:  Temperature is in 1/2 degree C increments, so the number needs to be divided in half on the report.

I was promised when I participated in the ERS56XX Beta the fan inlet temperature was going to be on the roadmap.  It never materialized, and I don't see it in the 4950 MIBs either.  We just got our first box of 59100's in, so I haven't had the chance to play with them to see if my wish has finally come true!

Fan Status (UP/Down)
1.3.6.1.4.1.45.1.6.3.3.1.1.10.6 Instance 10.0 through 15.0 for unit 1
1.3.6.1.4.1.45.1.6.3.3.1.1.10.6 instance 20.0 through 25.0 for unit 2
Pattern continues until you get to 80.0 through 85.0 for unit 8.

Power Supply (Up/Down)
1.3.6.1.4.1.45.1.6.3.3.1.1.10.4 instance 10.0 to 13.0 for unit 1 (Assuming 4 power supplies max)
Second switch starts at 20.0 - pattern is consistent with Fan Status.

Overall Switch Status (Up/Down)
.1.3.6.1.4.1.45.1.6.3.3.1.1.10.3.10 Instance 0 Unit 1
.1.3.6.1.4.1.45.1.6.3.3.1.1.10.3.20 Instance 0 unit 2
Continues until
.1.3.6.1.4.1.45.1.6.3.3.1.1.10.3.80 Instance 0 Unit 8

TankII

Offline Abdul

  • Jr. Member
  • **
  • Posts: 35
Re: PRTG CPU Monitoring
« Reply #2 on: July 11, 2017, 09:13:39 AM »
Thanks Tankll

Does it matter how many switches in a stack?
And what type of sensors did you use? Custom SNMP sensors or Custom Advanced SNMP Sensors?
« Last Edit: July 11, 2017, 09:19:17 AM by Abdul »

Offline TankII

  • Hero Member
  • *****
  • Posts: 544
Re: PRTG CPU Monitoring
« Reply #3 on: July 11, 2017, 01:03:15 PM »
If you see the patterns I listed with each OID, you can built it for up to eight switches in a stack.  Since ours rarely go above four, I only built it that far.

We enable SNMP v2c with a complex RO and RW string combination, and then use MRTG, WhatsUP, and other discovery/polling/graphing tools.  Make sure you label your ports, change the name of the switches, and configure logging too! 
Without knowing the tool in question, two things come to mind:
1)  Always use SNMP v2c or V3, since any graphing on SNMPv1 will loop the graph when the traffic hits 100Mbit
2)  Validate your MIBS using a free tool such as Getif so you don't get empty graphs.

TankII