This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

No MAC addresses or IP addresses show up on Cisco device ports

First of, I am pretty new to these products.

I have an issues in our organization.  No MAC addresses or IP addresses appear on ports for cisco devices entered into our NMS.  We are using UDT 3.0.2, NCM 7.2.2.  Other regions in our organization are not having the same problems with their cisco equipment.

Key differences

-We do not use vlan 1 as management Vlan

-We use snmp v3 (I reverted back to version 2, no difference)

I use a cisco 4507 as my core device.  L2 connection to all switches in my network.  All my Brocade switches do not have this problem. SNMP polling does not fail.  All other information has no problems being polled

I am sort of in a rush today, so I will add any information that is asked for, or as I remember and get a chance to.

What sort of MIB table is needed for all aspects to work correctly.  Is there a way I can get a checksheet for all components needed to get this functioning.  My Brocade switches are L3.  My cisco gear is a combo of L2 and L3, and no MAC address show up on any cisco switchports.  With the exception of thru trunk ports back to core.

I would be hugely appreciative if some one could walk me thru the process.  Ultimately, what I would love is a breakdown of how some aspects of the polling works.  I'm not sure if it's and architecture thing or not?

Please feel free to ask questions!

  • Wooly,

    does your 4507 contains any L3 info? ( or any other of the devices discovered by UDT?). Regarding: "No MAC addresses or IP addresses appear on ports for cisco devices entered into our NMS" - As there isn't integration between NCM and UDT on the level of discovered devices, you have to run separate discovery in UDT, then to choose from available ports etc.

  • I have SVI's on the 4507.  The gateway addresses for each vlan on our network.  We have access layer switches directly connected and a direct connection to another distribution 4507.  Show ip route shows subnets for the corresponding  vlans are "directly connected"   to the vlan numbers.  Devices that are plugged in to the chassis show IP information, mac addresses and Vlan info on the home tab.

    -When looking at ports on the home tab for my 4507 There are Vlan interfaces, and matching "unrouting vlan interfaces" that both have the same devices associated

    -If I was to use device tracker for a MAC address that is on a cisco access switch,  I will find the mac, ip, and computer name node port to be Vlan(#)  attached to 4507  connection type unknown

    -If I was to use device tracker for a MAC address that is on a brocade switch, I will find the two entrys for "NODE PORT OR SSID" one for the Vlan(#) and the next would be the actual switcport the device is connected to on the brocade, with the connection type "DIRECT"

    What do you mean exactly "there isn't integration between NCM and UDT on the level of discovered devices"  I'm just trying to understand better.

    I don't get why mac address do not at least show up on switchports of cisco switches when I am looking at the home tab "NCM" of the switch. It is sending out an "snmp get"  command to that switchport right?  How does that get lost in the shuffle.  I assume, maybe, the way it is designed to map IP's to MAC's, the software is reliant on the routing device being that switches are usaually an L2 device.  Is this how Device tracker and NCM work?

  • UDT NOT SHOWING IP ADDRESSES IN SWITCH PORT VIEW (this is entered for keyword searches)

    So we were having the same issue in our environment as well. I contacted SW for support on the EVAL edition. I have sent them the requested log information for our systems and the UDT Compatibility Checker logs as well. as for your IP and MAC issue its a SQL issue in our case where on the SQL server we had to change... here is the snippit from the email i recieved back from support...

    I hope this message finds you well. I got some feedback from our team on the UDT jobs not completing correctly. They mentioned the UDT Compatibility Checker output was not placed in diagnostics. Would you kindly generate those output files to send over to me to provide to the team. In reviewing the information in the diagnostics it seems they found the following:

    It looks like the SQL server connection properties has been configured to govern the query cost limit which is cancelling UDT queries.

    Kindly ask the customer to disable this setting and try running the Layer 3 job. The instructions to enable/disable and adjust the value of this setting is available in the below link preferably using SQL management studio

    http://technet.microsoft.com/en-us/library/ms191219.aspx.


    if your lazy (like me emoticons_happy.png ) here is the snippit from the link emoticons_happy.png


    Using SQL Server Management Studio


    To configure the query governor cost limit option

    1. In Object Explorer, right-click a server and select Properties.
    2. Click the Connections page.
    3. Select or clear the Use query governor to prevent long-running queries check box. If you select this check box, in the box below, enter a positive value, which the query governor uses to disallow execution of any query with a running length exceeding that value.
  • Having the same issue here and there.  It's very frustrating and we've upgraded to the most recent releases of everything.  It seems like it broke two or three upgrades ago.  I also tried the above fix but we didn't have it checked off to begin with.  By the looks of it, it seems quite a few others are having the same issue with the amount of view this post has.  If anyone can help or have any suggestions, I'd greatly appreciate it.

  • Hi,

    we're sorry you experiencing similar problems as some other users above. This is not an usual case and from our experience it requires investigation of the configuration of your environment. If you call/open a support ticket with us, we will help you to fix this. So far there is no universal answer to such problem. Our architecture team collects the similar feedback and already has designed an improved infrastructure that will prevent such cases from happening in the future.

  • I will go ahead and put in another ticket on the issue.  Thanks.

  • I maybe seeing the same thing.

    I can't get UDT to display switch port info on C2960s stacks in one IDF at a different IDF (same switch type) I'm not getting MAC, IP or switch port info.

    Sometimes I add a user and the host pops right in but no matter what I do I never get enough info to justify the money spent.

    Funny, the eval went fine using just my accounts on a local switch.

  • Four hours ago I disconnected a host on my UDT watch list and moved it to a different switch port.

    UDT is still showing the host being connected to the old port.
    Wondering is there was a firewall causing this I've disabled the firewall on my Solarwinds server and my target.

    Nothing!

  • This really annoying.

    I downloaded the evaluation product and it just worked.

    Directions were straightforward and I believed the product had value to the business.

    We decide to buy and apparently a newer release of UDT shipped, and that is what I installed and licensed but now it doesn't work.

    Out of my ten IDF's all having the same Cisco 2960s switches most stacked only one of them ever downloaded the ports and even the one that did has failed to see changes to my test port.UDT did see the device attached to a switch port but never even after three days see that the device was disconnected so according to Solarwinds its still there almost a week later.

    I'm at a point that opening a support ticket with Solarwinds is my only recourse as I've done everything recommended by others without improvement.

    One other concern I have is that we use a pair Cisco Nexus 5548 at the core that has very recent 7.3 firmware.

    As I've been trying to troubleshoot this mess I've read that others having Cisco Nexus have also reported problems.  

  • With the help of the wonderful Compatibility Checker tool I've been able to identify the problem (31016: Authorization Error) is when SNMPv3 authentication is set, SHA and privacy protection are AES256, and private mode is selected.

    The Cisco CCIE I have working on this so far has been unable to fix this using the Solarwinds recommendation's in the knowledge base.

    I know that others have had this or a similar problem so I'll let everyone know how it get's resolved.