Skip to main content
Open

Monitor physical links in a port-channel

Related products:Qapp
  • May 18, 2022
  • 2 replies
  • 31 views
C

When one physical link goes down inside a port-channel with mupltiple physical links, your port-channel will continue to work and unless you monitor the physical connections, there will be no alert. When the port-channel is used for a Cisco FEX (Fabric Extender) it will be difficult to model this in your monitoring because there is not endpoint to define it on.
My suggestion: collect all port-channels of a map and their physical interfaces and then query the status of the interfaces. Port-channels with less than 100% working interfaces are marked on the map.

This could also be used for pre and post change checks.

    2 replies

    prallu
    • prallu
    • Frequent Participant
    • 7 replies
    • August 4, 2022

    Hello Carsten,

    I will show you how we can achieve this using our new Network Intent(NI) 
    You need to follow below steps. 

    • You need to fetch all interfaces and if they are part of any port channels. 
    Parse All Interfaces and Port-Channels attached to them
    • In a second parser, you can fetch the interface statuses for each interface in the device
    • Parse all the interfaces and their statuses

       

    • Once we have above 2 details parsed, we have a capability in NI to create a compound table with the help of a Paired Key, in this case paired key will be the interface.

            

    Create a compound table with the use of Paired Keys

    After the creation of compound table - we need to create a diagnosis with the values we have. The diagnosis will basically look like this for this case. 

    Sample Diagnosis to check the status of interface part of a port channel

    Once we have the diagnosis created and run against the device, we will see alerts as below. 

    Alerts for a device with interfaces part of port-channel down

    NI will work for a single Device. To expand this to a group of devices or to a whole site, you can make use of the Network Intent Cluster. 

    NIC Framework

    1. Input Device - Full Site
    2. Seed NI - Parent NI we created earlier

    Seed NI is the NI that was previously created


    3. Seed Logic - Device Level Logic
    4. Device Classifiers - Cisco IOS Switch(for this example)
    5. Group Eigen Value - Hostname

    Group devices based on the Eigen Value


    6. Target Seed - True
    7. Create Member NIs - In Creation settings, select for a L2 Map to be created for the intent. 

     

    Finally run all the intents and you will get status as below. This NI is created to show only the problematic interfaces.

     

    One of 299 Intents part of this NIC has problems with Interfaces in a port-channel

     

    Hope this answers your questions and help you understand the new Network Intent Framework.
    Let us know if you have any questions or please feel free to raise a support ticket with your requirement. 

     

    Thanks,

    Pramod

    pallu
    C
    Forum|alt.badge.img+4
    • Carsten.Schmidt
    • Author
    • Brainy
    • 82 replies
    • August 5, 2022

    Thank you Prallu.
    Certainly this is one way of doing it but I was looking for a built-in solution instead of configuring intends for hours. Although I applaud NB’s capabilities of letting administrators build their own complex intends I was hoping that the information which NB collects is not simply regurgitated but analyzed to give admins information about config, status mismatches similar to the MTU/speed mismatch qapp.


    Follow NetBrain on LinkedIn

    Terms & ConditionsCookie settingsAccessibility statement