Monitoring IBM Server Hardware

Hello,

unfortunately we don't have any experience with this so far, but Google provided the following links:
ibmSystemMIB
ibmServeRAIDMIB
Configuring SNMP agent community information

As you might know, we just added support for Dell and HP servers to PRTG. IBM x86 server sensor are already in development, but we can't give you any estimates, when they will be released.

Any news about release date for IBM x86 server sensors ?

We do never give estimates to take the pressure off. All we can tell you is that the IBM System X sensors will be part auf our stable 13.2.4 release. The feature development is already done for this version. Now, our qa department ist testing it and we will only do bugfixing, if neccessary. This version already took much longer than we had hoped, but we are confident, that it will be released really soon.

The new sensors for IBM System x are "active" ! But i can find them only via Automatic Search. If i'd like to manually add a "system x" sensor, i can't find it in the add dialog. Is this a bug ?

Please see this post: add new sensor types

Hello,

I tested your x3630M4, x3650M4 and x3550M4 probes on a server and it does not work correctly.

I use an IMM network port snmpv3

It works for: SNMP IBM System X System Health Sensor (TEMP, VOLTAGE AND FAN)

It will not work for: Physical Memory Sensor error (code: PE187) Physical Disk Sensor error (code: PE188) Logical Disk Sensor error (code: PE188)

Do you have a solution?

Can you please try running a scan with the SNMP tester to see what the device is reporting for SNMP? Perform a walk on the OID 1.3.6.1.4.1.2.3.51.3.1 to do this for the IBM IMM.

25/10/2013 17:38:38 (2 ms) : SNMP V3
25/10/2013 17:38:38 (2 ms) : Custom OID 1.3.6.1.4.1.2.3.51.3.1
25/10/2013 17:38:39 (273 ms) : -------
25/10/2013 17:38:39 (274 ms) : Value: No such object (SNMP error # 222)
25/10/2013 17:38:39 (275 ms) : Done

When you did this scan you used the custom OID field. Please use the walk field and please also try walking this OID 1.3.6.1.4.1.2.6.159.1.1.80.

Uptime :

----------------------- New Test -----------------------
Paessler SNMP Tester 5.1
25/10/2013 17:53:30 (4 ms) : Device: 192.168.40.33
25/10/2013 17:53:30 (6 ms) : SNMP V3
25/10/2013 17:53:30 (9 ms) : Uptime
25/10/2013 17:53:30 (98 ms) : -------
25/10/2013 17:53:30 (100 ms) : Value: 1717
25/10/2013 17:53:30 (102 ms) : Done

OID : 1.3.6.1.4.1.2.6.159.1.1.80

----------------------- New Test -----------------------
Paessler SNMP Tester 5.1
25/10/2013 17:52:06 (3 ms) : Device: 192.168.40.33
25/10/2013 17:52:06 (4 ms) : SNMP V3
25/10/2013 17:52:06 (6 ms) : Custom OID 1.3.6.1.4.1.2.6.159.1.1.80
25/10/2013 17:52:06 (133 ms) : -------
25/10/2013 17:52:06 (134 ms) : Value: No such object (SNMP error # 222)
25/10/2013 17:52:06 (136 ms) : Done

We are also having a hard time monitoring IBM hardware over IMM. We have setup our SNMP correctly, and we have tried using both built-in sensors from PRTG and downloaded IBM MIBs, but we can't seem to get a status of the hard drives/Raid.

If I pull one disk out from the RAID, I can via the SNMP tester test OID .1.3.6.1.4.1.2.3.51.3.1.4.2.1.3.2, and I get the error: Value: The Drive "Drive 1" has been disabled due to a detected fault

This is all fine, however If the hard dives/raid is OK, I can't test the OID.

What I want (correct me if it's not possible, please). We have our PRTG Core. We install Remote Probes on our customers ADMIN/Monitor server. We use the Remote Probe server to poll the other servers. When a hard drive fails on a server, we want the Remote Probe to report back to our PRTG Core server, and a “red status” will appear on our Maps and e-mails will be sent out.

I hope you understand what I'm trying to describe

@PMAdmin you need to run a walk on that OID since the custom OID will not work. This OID is a base OID for the tree so there will not be any values, the values are for the OIDs underneath that base.(See Screenshot)

@Annfinn - If that OID for the RAID status only shows when there is a failure, then there isn't any way to monitor this through PRTG. The problem here is that when you set up a custom OID in PRTG it needs to always be there either saying OK or Bad or something like that. If the OID disappears and re-appears, this cannot work. We can look into possibly creating a sensor that would scan that whole tree, since this tree is essentially a "when things fail OIDs show here" tree but this is something that would need to be voted up and requested before we can invest the time in it.

The only way this may be possible would be to create the sensor that would be in a down state when the RAID is ok and in an OK state when it's down and then use a sensor factory sensor to invert the status but this is quite a work-around.

Hello Greg,

Thanks for the reply.

Point me to the voting please :)

How would you suggest we monitor Physical disks and Raids over IMM (IBMs Integrated Management Module)? It's essential that we are able to monitor the Physical state of the server, and the logical only gets us so far. :)

To put this on the wishlist you can create a new thread with the Title: Wishlist IMM Fault Tree Sensor or something like that and if others see this and want to vote this feature up they can vote there.

From looking over what is publicly available for MIBs concerning the IMM it does not look like monitoring disks is possible. You would likely need to install the IBM System Director Agent on the system to be able to pick up this information through PRTG.

There may be some OIDs that I cannot see without having an IBM account that you may be able to monitor for HDD status through the IMM but that is something you would need to find on IBM's website or by talking with their tech support. You could then use a Custom SNMP sensor or an SNMP library sensor to monitor those stats.

Hello again,

Thanks for the reply.

I've tried installing IBM System Director Common Agent and Platform Agent, without any luck getting the status of the Disks. :(

I understand that this might ultimately be IBM's problem; however I was hoping that there would be something we could do PRTG-wise. :(

The IMM is able to see the status of the 'degraded' Raid and I'm picking up the OID as long as the RAID, or rather disk is in that state.

What OS did you install the Systems Director agent onto? Can you try doing a walk on the base OID 1.3.6.1 and then send over the results in a ticket? You can just attach the output on an email and send it to [email protected] and reference this KB article.

The Agent is installed a Windows Server 2008 R2. I will send the OID walk from 2 or 3 servers ASAP.

Hi

On the IBM System Health Sensors for fans : '''' is not a valid integer value and other sensors are not working (hdd physical and logical disk)

20/11/2013 00:57:19 (5426 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.1.1 = "1"
20/11/2013 00:57:19 (5442 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.1.2 = "2"
20/11/2013 00:57:19 (5458 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.1.3 = "3"
20/11/2013 00:57:19 (5474 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.1.4 = "4"
20/11/2013 00:57:19 (5490 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.1.5 = "5"
20/11/2013 00:57:19 (5505 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.1.6 = "6"
20/11/2013 00:57:19 (5521 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.1.7 = "7"
20/11/2013 00:57:19 (5536 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.1.8 = "8"
20/11/2013 00:57:19 (5558 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.2.1 = "Fan 1A Tach"
20/11/2013 00:57:19 (5573 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.2.2 = "Fan 1B Tach"
20/11/2013 00:57:19 (5590 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.2.3 = "Fan 2A Tach"
20/11/2013 00:57:20 (5607 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.2.4 = "Fan 2B Tach"
20/11/2013 00:57:20 (5624 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.2.5 = "Fan 3A Tach"
20/11/2013 00:57:20 (5640 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.2.6 = "Fan 3B Tach"
20/11/2013 00:57:20 (5656 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.2.7 = "Fan 4A Tach"
20/11/2013 00:57:20 (5672 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.2.8 = "Fan 4B Tach"
20/11/2013 00:57:20 (5688 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.3.1 = " 34% of maximum"
20/11/2013 00:57:20 (5704 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.3.2 = " 30% of maximum"
20/11/2013 00:57:20 (5720 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.3.3 = " 34% of maximum"
20/11/2013 00:57:20 (5736 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.3.4 = " 31% of maximum"
20/11/2013 00:57:20 (5751 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.3.5 = " 34% of maximum"
20/11/2013 00:57:20 (5767 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.3.6 = " 31% of maximum"
20/11/2013 00:57:20 (5783 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.3.7 = "offline"
20/11/2013 00:57:20 (5800 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.3.8 = "offline"
20/11/2013 00:57:20 (5815 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.4.1 = "0"
20/11/2013 00:57:20 (5832 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.4.2 = "0"
20/11/2013 00:57:20 (5848 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.4.3 = "0"
20/11/2013 00:57:20 (5864 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.4.4 = "0"
20/11/2013 00:57:20 (5880 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.4.5 = "0"
20/11/2013 00:57:20 (5895 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.4.6 = "0"
20/11/2013 00:57:20 (5911 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.4.7 = "0"
20/11/2013 00:57:20 (5928 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.4.8 = "0"
20/11/2013 00:57:20 (5943 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.5.1 = "0"
20/11/2013 00:57:20 (5959 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.5.2 = "0"
20/11/2013 00:57:20 (5977 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.5.3 = "0"
20/11/2013 00:57:20 (5993 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.5.4 = "0"
20/11/2013 00:57:20 (6009 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.5.5 = "0"
20/11/2013 00:57:20 (6025 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.5.6 = "0"
20/11/2013 00:57:20 (6042 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.5.7 = "0"
20/11/2013 00:57:20 (6058 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.5.8 = "0"
20/11/2013 00:57:20 (6074 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.6.1 = "0"
20/11/2013 00:57:20 (6089 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.6.2 = "0"
20/11/2013 00:57:20 (6106 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.6.3 = "0"
20/11/2013 00:57:20 (6123 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.6.4 = "0"
20/11/2013 00:57:20 (6140 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.6.5 = "0"
20/11/2013 00:57:20 (6158 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.6.6 = "0"
20/11/2013 00:57:20 (6174 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.6.7 = "0"
20/11/2013 00:57:20 (6190 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.6.8 = "0"
20/11/2013 00:57:20 (6213 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.7.1 = "0"
20/11/2013 00:57:20 (6229 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.7.2 = "0"
20/11/2013 00:57:20 (6244 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.7.3 = "0"
20/11/2013 00:57:20 (6260 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.7.4 = "0"
20/11/2013 00:57:20 (6276 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.7.5 = "0"
20/11/2013 00:57:20 (6293 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.7.6 = "0"
20/11/2013 00:57:20 (6309 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.7.7 = "0"
20/11/2013 00:57:20 (6325 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.7.8 = "0"
20/11/2013 00:57:20 (6342 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.8.1 = "962"
20/11/2013 00:57:20 (6357 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.8.2 = "960"
20/11/2013 00:57:20 (6374 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.8.3 = "962"
20/11/2013 00:57:20 (6390 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.8.4 = "960"
20/11/2013 00:57:20 (6406 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.8.5 = "962"
20/11/2013 00:57:20 (6421 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.8.6 = "960"
20/11/2013 00:57:20 (6437 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.8.7 = "0"
20/11/2013 00:57:20 (6453 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.8.8 = "0"
20/11/2013 00:57:20 (6469 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.9.1 = "0"
20/11/2013 00:57:20 (6484 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.9.2 = "0"
20/11/2013 00:57:20 (6500 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.9.3 = "0"
20/11/2013 00:57:20 (6515 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.9.4 = "0"
20/11/2013 00:57:20 (6531 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.9.5 = "0"
20/11/2013 00:57:20 (6547 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.9.6 = "0"
20/11/2013 00:57:20 (6563 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.9.7 = "0"
20/11/2013 00:57:21 (6695 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.9.8 = "0"
20/11/2013 00:57:21 (6712 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.10.1 = "Normal"
20/11/2013 00:57:21 (6727 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.10.2 = "Normal"
20/11/2013 00:57:21 (6743 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.10.3 = "Normal"
20/11/2013 00:57:21 (6759 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.10.4 = "Normal"
20/11/2013 00:57:21 (6776 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.10.5 = "Normal"
20/11/2013 00:57:21 (6791 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.10.6 = "Normal"
20/11/2013 00:57:21 (6807 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.10.7 = "Unknown"
20/11/2013 00:57:21 (6823 ms) : 1.3.6.1.4.1.2.3.51.3.1.3.2.1.10.8 = "Unknown"

The reason that the sensor is returning that error is because the server is returning

1.3.6.1.4.1.2.3.51.3.1.3.2.1.2.1 = "Fan 1A Tach"
1.3.6.1.4.1.2.3.51.3.1.3.2.1.2.2 = "Fan 1B Tach"
1.3.6.1.4.1.2.3.51.3.1.3.2.1.2.3 = "Fan 2A Tach"
1.3.6.1.4.1.2.3.51.3.1.3.2.1.2.4 = "Fan 2B Tach"
1.3.6.1.4.1.2.3.51.3.1.3.2.1.2.5 = "Fan 3A Tach"
1.3.6.1.4.1.2.3.51.3.1.3.2.1.2.6 = "Fan 3B Tach"
1.3.6.1.4.1.2.3.51.3.1.3.2.1.2.7 = "Fan 4A Tach"
1.3.6.1.4.1.2.3.51.3.1.3.2.1.2.8 = "Fan 4B Tach"
1.3.6.1.4.1.2.3.51.3.1.3.2.1.3.1 = " 34% of maximum"
1.3.6.1.4.1.2.3.51.3.1.3.2.1.3.2 = " 30% of maximum"
1.3.6.1.4.1.2.3.51.3.1.3.2.1.3.3 = " 34% of maximum"
1.3.6.1.4.1.2.3.51.3.1.3.2.1.3.4 = " 31% of maximum"
1.3.6.1.4.1.2.3.51.3.1.3.2.1.3.5 = " 34% of maximum"
1.3.6.1.4.1.2.3.51.3.1.3.2.1.3.6 = " 31% of maximum"
1.3.6.1.4.1.2.3.51.3.1.3.2.1.3.7 = "offline"
1.3.6.1.4.1.2.3.51.3.1.3.2.1.3.8 = "offline"

These are last two values are strings which the sensor is not expecting. I have submitted a bug ticket for this and we will try and get this solved as soon as possible.

Also, the other issue here is that the OID's that are used for the other sensors that you mentioned such as the physical and logical disks are not here. This is because the IMM does not support the monitoring of these components. What you need to do is install the IBM System Director agent on the OS of the server and enable SNMP so that the server will then also be able to respond to the OID's that are needed to monitor the HDDs.

Excactly Greg - as I told you in the e-mails, we've got it working by installing the agent AND configurating the SNMP service on Windows. Did you update the Knowledgebase as per request?

Next stop: Citrix hardware monitoring on IBM hardware! ;)

I updated this a few weeks ago in the article that we wrote about the requirements for monitoring IBM servers.

Glad that this is all working!!

Hmmm.... IBM System Director 6.3 (latest Verison) is out of Service since April 2018 what to do now?

Hi there,

I'm afraid that I could not find any updated version of the System Director Tool, which might mean that these systems are deprecated by IBM due to their age? Using the old IBM System Director should still work, if you need a new version which might not be given out publicly, I recommend to get in contact with IBM directly, they will know best.

Best regards, Felix