About Alarm ID

[MARUBUN SUPPORT]

Hi Team,

Our customer is executing Spectrum on a Linux server(SystemEDGE for LINUX), and the alarm IDs that occurred during execution were as follows:

• When the I/F was put into admin down state by executing ip link set and I/F name down, the following Alarm ID was returned.
      Alarm ID ： 0x1040b / LINK DISABLED
• CPU usage exceeds threshold
      Alarm ID ： 0x10f03 / High CPU utilization (*1)
• Memory usage exceeds threshold
      Alarm ID ： 0x10f06 / High Memory Utilization (*1)

*1 : I was unable to find this Alarm ID in the Spectrum manual.

Q1:
Please let me know if I'm misinterpreting the above Alarm IDs.

Q2:
Please tell me if these Alarm IDs change depending on the type of model or situation.

Q3:
I think the cause code for the I/F status will change depending on whether it went down as shown below.
Please tell me if the Alarm ID changes depending on the situation when it went down.

0x10409: Administrative shutdown (*2)
0x1040a: Bad link detected

* 2 : I was unable to find this Alarm ID in the Spectrum manual.
       Please let me know if there is any mistake in my interpretation of this Alarm ID.

Best Regards,
Marubun Support

[Joseph Ackley]

Maruban,

Please see my answers below.

Q1:
Please let me know if I'm misinterpreting the above Alarm IDs.
ANSWER: You are not misinterpreting the above Alarm ID's. 

Q2:
Please tell me if these Alarm IDs change depending on the type of model or situation.
ANSWER: No they don't change depending on the type of model. They are global.

The LINK DISABLED is asserted when the ifAdminStatus of the interface is set to DOWN on an interfaceif PollPortStatus is set to "Yes/True" or that is part of a connection where Live Links/Pipes is enabled. The associated event id 0x10d12 and the pcause id 0x1040b are mentioned in the "Port Status Events and Alarms" section of the documentation but that is about it.

For the 0x10f03 and 0x10f06, you are correct that there is no document reference to these alarm ids. However, please review the "Calculating CPU and Memory Utilization" section of the documentation to get an idea on how CPU and memory thresholding is done in Spectrum.

Q3:
I think the cause code for the I/F status will change depending on whether it went down as shown below.
Please tell me if the Alarm ID changes depending on the situation when it went down.

0x10409: Administrative shutdown (*2)
ANSWER: I cannot find any reference for this as well so I am not sure when this is asserted.

0x1040a: Bad link detected
ANSWER: This alarm is asserted if ifAdminStatus is up AND ifOperStatus is down AND port status alarms are enabled. Please reference the "How to Configure Port/Interface alarming in DX NetOps Spectrum when a port/interface model goes down?" knowledge document for more information.

Hope this helps.

Joe