NetXMS Support Forum

English Support => General Support => Topic started by: jcallan017 on June 25, 2025, 02:53:00 AM

Title: Unable to auto discover certain Ruckus ICX Switch nodes
Post by: jcallan017 on June 25, 2025, 02:53:00 AM
I am trying to add a group of nodes by active discovery of their subnet, whilst I can add the nodes manually and they pass all the discovery filters and the snmp strings pass, it fails to add on poll.discovery AcceptNewNodeStage2, see below log file:

I've changed/removed IP's or strings for sanity.

2025.06.25 09:26:36.825 *D* [poll.discovery    ] AcceptNewNodeStage1(A.B.C.D): MAC address not found
2025.06.25 09:26:36.825 *D* [poll.discovery    ] AcceptNewNodeStage1(A.B.C.D): filter flags=0303
2025.06.25 09:26:36.825 *D* [poll.discovery    ] AcceptNewNodeStage1(A.B.C.D): filter - checking address range
2025.06.25 09:26:36.826 *D* [poll.discovery    ] AcceptNewNodeStage1(A.B.C.D): filter - range check result is true
2025.06.25 09:26:36.828 *D* [agent.conn.450171  ] New connection created (address=A.B.C.D port=4700 compression=allowed)
2025.06.25 09:26:36.830 *D* [agent.conn.450171  ] Cannot establish connection with agent at A.B.C.D:4700
2025.06.25 09:26:36.830 *D* [poll.discovery    ] HostIsReachable(A.B.C.D): agent connection check failed with error 910 (Connect failed)
2025.06.25 09:26:36.830 *D* [snmp.discovery    ] SnmpCheckCommSettings(A.B.C.D): starting check (proxy=0, originalPort=0)
2025.06.25 09:26:36.830 *D* [snmp.discovery    ] SnmpCheckCommSettings(A.B.C.D): checking port 161
2025.06.25 09:26:36.831 *D* [snmp.discovery    ] SnmpCheckV3CommSettings(A.B.C.D): trying
2025.06.25 09:26:36.845 *D* [snmp.discovery    ] SnmpCheckV3CommSettings(A.B.C.D): trying
2025.06.25 09:26:36.854 *D* [snmp.discovery    ] SnmpCheckV3CommSettings(A.B.C.D): trying
2025.06.25 09:26:36.868 *D* [snmp.discovery    ] SnmpCheckV3CommSettings(A.B.C.D): trying
2025.06.25 09:26:36.879 *D* [snmp.discovery    ] SnmpCheckV3CommSettings(A.B.C.D): trying
2025.06.25 09:26:36.894 *D* [snmp.discovery    ] SnmpCheckV3CommSettings(A.B.C.D): failed
2025.06.25 09:26:36.894 *D* [snmp.discovery    ] SnmpCheckCommSettings(A.B.C.D): trying version 1 community 'removed'
2025.06.25 09:26:36.898 *D* [snmp.discovery    ] SnmpCheckCommSettings(A.B.C.D): success (version=1)
2025.06.25 09:26:36.898 *D* [poll.discovery    ] HostIsReachable(A.B.C.D): SSH connection probe disabled
2025.06.25 09:26:36.903 *D* [ndd                ] FindDriverForNode(A.B.C.D): found potential device driver PING3 with priority 63
2025.06.25 09:26:36.907 *D* [poll.discovery    ] AcceptNewNodeStage1(A.B.C.D): selected device driver GENERIC
2025.06.25 09:26:38.229 *D* [ndd.common        ] NetworkDeviceDriver::getInterfaces(0x7f7b1024fe00): no prefix reference for A.B.C.D
2025.06.25 09:26:38.281 *D* [poll.discovery    ] DiscoveredAddressPoller: stage 2 processing for address A.B.C.D/24 in zone 0 (source type Active Discovery, source node
2025.06.25 09:26:38.282 *D* [poll.discovery    ] AcceptNewNodeStage2(A.B.C.D): interface list available, checking for duplicates
2025.06.25 09:26:38.282 *D* [poll.discovery    ] AcceptNewNodeStage2(A.B.C.D): IP address is a broadcast address for subnet 0.0.0.0/0
2025.06.25 09:26:38.282 *D* [poll.discovery    ] ProcessDiscoveredAddressStage2(A.B.C.D): address discarded

OS - AlmaLinux release 9.5 (Teal Serval)
NetXMS Version - 5.2.0
Node Device - Ruckus ICX7150-24-POE
Node Version - Version 08.0.90jT211
I can add this node manually without issue and other nodes in the same discovery subnet add via discovery poll, it's just nodes that meet this criteria that fail.

snmpwalk from the server gives the following:
snmpwalk -v 2c -c removed  A.B.C.D 1.3.6.1.2.1.2.2.1.2
IF-MIB::ifDescr.1 = STRING: GigabitEthernet1/1/1
IF-MIB::ifDescr.2 = STRING: GigabitEthernet1/1/2
IF-MIB::ifDescr.3 = STRING: GigabitEthernet1/1/3
IF-MIB::ifDescr.4 = STRING: GigabitEthernet1/1/4
IF-MIB::ifDescr.5 = STRING: GigabitEthernet1/1/5
IF-MIB::ifDescr.6 = STRING: GigabitEthernet1/1/6
IF-MIB::ifDescr.7 = STRING: GigabitEthernet1/1/7
IF-MIB::ifDescr.8 = STRING: GigabitEthernet1/1/8
IF-MIB::ifDescr.9 = STRING: GigabitEthernet1/1/9
IF-MIB::ifDescr.10 = STRING: GigabitEthernet1/1/10
IF-MIB::ifDescr.11 = STRING: GigabitEthernet1/1/11
IF-MIB::ifDescr.12 = STRING: GigabitEthernet1/1/12
IF-MIB::ifDescr.13 = STRING: GigabitEthernet1/1/13
IF-MIB::ifDescr.14 = STRING: GigabitEthernet1/1/14
IF-MIB::ifDescr.15 = STRING: GigabitEthernet1/1/15
IF-MIB::ifDescr.16 = STRING: GigabitEthernet1/1/16
IF-MIB::ifDescr.17 = STRING: GigabitEthernet1/1/17
IF-MIB::ifDescr.18 = STRING: GigabitEthernet1/1/18
IF-MIB::ifDescr.19 = STRING: GigabitEthernet1/1/19
IF-MIB::ifDescr.20 = STRING: GigabitEthernet1/1/20
IF-MIB::ifDescr.21 = STRING: GigabitEthernet1/1/21
IF-MIB::ifDescr.22 = STRING: GigabitEthernet1/1/22
IF-MIB::ifDescr.23 = STRING: GigabitEthernet1/1/23
IF-MIB::ifDescr.24 = STRING: GigabitEthernet1/1/24
IF-MIB::ifDescr.25 = STRING: Management
IF-MIB::ifDescr.65 = STRING: GigabitEthernet1/2/1
IF-MIB::ifDescr.66 = STRING: GigabitEthernet1/2/2
IF-MIB::ifDescr.129 = STRING: 10GigabitEthernet1/3/1
IF-MIB::ifDescr.130 = STRING: 10GigabitEthernet1/3/2
IF-MIB::ifDescr.131 = STRING: 10GigabitEthernet1/3/3
IF-MIB::ifDescr.132 = STRING: 10GigabitEthernet1/3/4

snmpwalk -v 2c -c removed A.B.C.D 1.3.6.1.2.1.4.20.1.2
IP-MIB::ipAdEntIfIndex = No Such Instance currently exists at this OID

snmpwalk -v 2c -c removed A.B.C.D 1.3.6.1.2.1.4.20.1.3
IP-MIB::ipAdEntNetMask = No Such Instance currently exists at this OID
It looks like this specific device or firmware is not using ip-mib or it's filtered?

Checking the ndd folder, there is no specific Brocade/Ruckus ndd file. Is this something you have come across before or have a potential workaround for?

Note, we have thousands of these devices in the field and have not come across it on other switch types or versions.
Title: Re: Unable to auto discover certain Ruckus ICX Switch nodes
Post by: Filipp Sudanov on June 25, 2025, 04:17:44 PM
Please check if you have subnet 0.0.0.0/0 under Entire Network. If it's not visible there, you pls login using the system user.

If it's there, pls delete it (make sure that nodes under are also present under other subnets or containers, otherwise they'll get deleted).
Title: Re: Unable to auto discover certain Ruckus ICX Switch nodes
Post by: jcallan017 on June 26, 2025, 05:23:07 AM
Hi Filipp,

Thanks for the quick response, I have checked under both my user and the sytem user and the 0.0.0.0 subnet does not exist. I have also checked via nxadm -c "show objects" for any match on 0.0.0.0 and nothing matches apart from a 10.0.0.0/24 network, so this doesn't look to be the cause of the issue, unless I am missing something.

Thanks,
Title: Re: Unable to auto discover certain Ruckus ICX Switch nodes
Post by: Filipp Sudanov on June 26, 2025, 07:46:42 AM
Let's double-check with script to be sure. Pls run this on "Entire network" object:

for (c : $object.children) {
  if (classof(c) == "Subnet")
    println(f"{c.name}: {c.ipAddr}/{c.ipNetMask}");
  else
    for (c2 : c.children) {
      println(f"{c2.name}: {c2.ipAddr}/{c2.ipNetMask}");

    }
}

it will print name of subnet and actual IP/mask (which can differ from subnet name)
Title: Re: Unable to auto discover certain Ruckus ICX Switch nodes
Post by: jcallan017 on June 27, 2025, 09:18:22 AM
Hi Filipp,

I ran this on the entire network and the only match for 0.0.0.0 was 10.0.0.0/24: 10.0.0.0/24.

Thanks,
Title: Re: Unable to auto discover certain Ruckus ICX Switch nodes
Post by: Filipp Sudanov on July 09, 2025, 12:00:09 AM
Hi, 

A few changes were made in the code, including better logging and special handling if 0.0.0.0/0 subnet was received during network discovery. This will be included in next patch release that should be in the coming days.