1819545 Members
2766 Online
109603 Solutions
New Discussion

incoherent values for alarms after upgrading from E0705P04 to E0705P10

 
Lewis1
Advisor

incoherent values for alarms after upgrading from E0705P04 to E0705P10

Hello,

Since upgrading from E0705P04 to E0705P10, I have some alarms which are not coherent for 4 devices (500 devices in total collected by imc):

 

interface_alert.png

These devices are 3 Cisco switchs and one HP 5130 switch. I tried deleting/re-adding these devices but the alarms reappears after some time. For the other devices, the values are coherent.

I did a rollback on my VM and tried to upgrade from E0705P04 to E0705P07, but I have the same problem, and I have no problem with E0705P04.

I don't know where I should look to troubleshoot this problem, does anyone have an idea ?

Thank you for your help.

Regards

4 REPLIES 4
jguse
HPE Pro

Re: incoherent values for alarms after upgrading from E0705P04 to E0705P10

Hello,

Please check the speed of the interfaces that are affected here as detected by iMC, via their Device Details -> Interface List -> open the affected interface (like Ten-GigabitEthernet1/0/51). Is the Speed (bps) shown as 10G, or something else?

Based on the alarm, it sounds like iMC is not detecting the interface bandwidth correctly, as it claims the "800Mbps" threshold is exceeded, which would be the default Threshold 2 for 1G interfaces, and not 10G.

For interface-specific indexes like the "Interface Receiving Rate" you showed, the alarm thresholds are variable depending on the speed detected for each interface - so a 1G interface would have a different alarm threshold than a 10G interface and so on.

If you use aggregations and thus have other speeds like 2G, 20G etc., you would have to manually define thresholds for alarming. This can be done via Performance Management -> Global Index Settings -> expand the interface index to modify, select the gear icon under Operation column, and select Modify Index. A window will appear like the following, where you can manually adjust each of the separate interface speeds and their thresholds:

IMC-Threshold.png

If the interface speed is detected incorrectly by iMC, you can check whether it's an iMC or device issue by polling the device's standard interface table via SNMP (for example using IMC's MIB Management tool), and looking at what speed is reported for the interface having issues.

 

Name: ifTable
Type: OBJECT-TYPE
OID: 1.3.6.1.2.1.2.2
Full Path: iso.org.dod.internet.mgmt.mib-2.interfaces.ifTable
Module: RFC1213-MIB
Parent: interfaces
Max Access: not-accessible
ComposedSyntax: SEQUENCE OF IfEntry
Status: mandatory
Description: A list of interface entries. The number of entries is given by the value of ifNumber.

 

Using MIB Management, you could right-click on ifTable and select Table View. This will show you all the values for the interfaces that the device reports in a nice table view, including the speed:
IMC-MIB Management.png
IMC should use whatever value the device provides for interface speed as reported above, so if this table shows an incorrect speed value for the interface, it's an issue on the device side.
 
Hope that helps.
Best regards,
Justin

Working @ HPE
Accept or Kudo
Lewis1
Advisor

Re: incoherent values for alarms after upgrading from E0705P04 to E0705P10

Thank you for your answer !

Speed is not the only index I have problem with, also broadcast per seconds display totally incoherent values.

We use a SFP (1G) adapter on this 10G port so IMC is detecting the speed correctly:

On interface list:

speed on interface listspeed on interface list

On the MIB browser:speed on mib browserspeed on mib browser

Sometimes, i have also incoherent values for the broadcast per second counters:

imc_bcast.png

Hope that helps you understanding this problem

jguse
HPE Pro

Re: incoherent values for alarms after upgrading from E0705P04 to E0705P10

Hello,

You had mentioned a Ten1/0/51 interface having this issue with speed. Is that the same interface Ten1/0/51 you are showing above in the MIB Management output? I'm wondering because the speed of that interface is showing as "4294967295" which doesn't seem right, even though the interface is down.

However since you mentioned the issue only happens after you upgraded iMC, you could also be encounting a new bug or otherwise something specific to your IMC installation. I think you will need to open a support case for this, so that we can investigate it directly.

Best regards,
Justin

Working @ HPE
Accept or Kudo
Lewis1
Advisor

Re: incoherent values for alarms after upgrading from E0705P04 to E0705P10

Sorry if I was not clear, I selected another interface of another equipment which had the same problem, but this adds confusion. On my previous screenshots, alerts were on the Ten1/0/52 interface, but its speed was correctly detected as you can see.

Anyway I will try to open a case for this problem because it seems weird.

Thank you for your help !