9 Replies Latest reply on Mar 22, 2012 7:39 AM by Dogeron

    Subnet scans

    twing

      I am using IPAM to monitor both DHCP subnets and manually entered subnets. I have only had it setup for a week or 2 but all the sudden all my scans time out and reschedule. Ive tried what I know (SNMP credentials, admin credentials are all correct) and am new to this tool. Any ideas were to start.

       

       

      Thanks

       

      Tim

        • Re: Subnet scans
          mavturner

          Are you evaluating or have you  purchased? If you have purchased, I would recommend opening a support case as I have not seen this before.

          Where do you see the information about the scans timing out? In the event log? How large are your subnets in general? Are you scanning large /8s or standard /24s? Is it just the DHCP jobs or the normal subnets scan (ICMP and SNMP scans against endpoints)?

            • Re: Subnet scans
              twing

              Im scanning /24's (but alot of them) and its using both ICMP or SNMP...but this just started happening

               

              We have purchased.

                • Re: Subnet scans
                  mavturner

                  Go ahead and open a support case. When you do, link to this thwack forum.

                  • Re: Subnet scans
                    Dogeron

                    Following a recent upgrade to the latest version (IPAM 2.0.1), I am now suffering a similar fate.

                    It appears that there is a hard-code 10 minute timeout on IPAM subnet scans that you can't get to.

                    And once the subnet gets timed-out the system puts it back into the queue again, which doesn't seem an entirely logical thing to do, as it's likely to continue to timeout. Worse, when it gets put back in the queue it gets prioritised (i.e. its raised to the next job) meaning that you get grid locked once you have 2 failed jobs in succession. Only manual intervention, canceling looping failed jobs will help. You can't re-schedule the job nor (apart from manually failing it) can you send it to the bottom of the queue....grrrrrr.

                    I've played with the settings and by un-checking "Enable SNMP neighbor scanning" together with reducing the number of scans to 1, and reducing the "Pings per address" to 3  I seem to be getting some subnets through again.

                    The problem seems to manifests itself with larger subnets, slow devices or a large, highly utilised device (like a stack of 7 Cisco 3750's) .

                    I've been given a suggestion that breaking the bigger subnet up into smaller logical groups within IPAM might work.

                    For me it's the next logical thing to try - it's not ideal but would be better than nothing.

                    Good luck,

                    Dog