HFC Nodes Monitoring | docsis.org

You are here

HFC Nodes Monitoring

5 posts / 0 new
Last post
hassaan
HFC Nodes Monitoring

How can an HFC node be monitored which doesn't support IP/SNMP? I'm trying to configure an SNMP based monitoring system to generate alarms whenever a node goes down but the problem is that the nodes are unmanaged -- they don't support IP/SNMP.
When a node goes down, reverse path fiber alarm in the form of red LED is received on ARRIS CHP Max5000 headend chassis which supports IP/SNMP.

mbowe
CHP MAX SNMP

Yep, you can poll the CHP return receivers for alarms.

For example we have 2RRX modules and we poll like this :

snmpget -v1 -c public x.x.x.x SCTE-HMS-HE-OPTICAL-RECEIVER-MIB ::heOpRxInputStatus.AAAABB.C

Where :
X.X.X.X is stack IP
AAAA is the chassis ID eg 1823
BB is the slot ID eg 08
C is the 2RRX input 1 or 2 eg 2

results :
1 = optical levels ok
2 = alarm (eg node has lost power, fiber cut etc)

hassaan
Thanks mbowe

That is what I'm trying to do on SolarWinds Orion Network Performance Monitor 9.1 but haven't been able to configure it so far.

I've configured Orion to receive traps from CHP. A sample trap is pasted below.

scteHmsTree.11.2.1.1.1.2.3.1.9.1736 =
scteHmsTree.11.2.1.1.1.2.3.1.8.1736 = CHP-2RRX
scteHmsTree.11.2.1.1.1.2.3.1.7.1736 = 1.3.6.1.4.1.2907.20.1.1.39.2
scteHmsTree.11.2.1.1.1.2.3.1.6.1736 = Rx B Optical Input Power Alarm, Alarm Cleared
scteHmsTree.11.2.1.1.1.2.3.1.5.1736 = B+ALFQwRCgA=
scteHmsTree.11.2.1.1.1.2.3.1.4.1736 = 1
scteHmsTree.11.2.1.1.1.2.3.1.3.1736 = -68
scteHmsTree.11.2.1.1.1.2.3.1.2.1736 = 1.3.6.1.4.1.5591.1.11.1.2.1.1.1.1.2.1679902.2
snmpTrapOID = SCTE-ROOT:scteHmsTree.0.5
sysUpTime = 8 days 19 hours 9 minutes 2.85 seconds

The alert generation part is where I'm stuck. How to generate alert for a particular node?

Msarmento
Try using indirect monitoring

When the node goes down, the immediate effect can be seen on the CMTS, the cables will go down too.
You can poll the total/registered modems for individual interfaces or MAC domain from the CMTS using SNMP. The object ids to poll vary from vendor, but every one has some sort of monitoring of this type.
Another suggestion is to associate a cable modem with the node and/or network element you want to monitor. If that modems goes down, it can indicate that there is a problem with the associated element.

Regards,

MS

hassaan
Thanks Msarmento

That is how monitoring is being done at the moment. On some CMTS ports, there are more than one node; so modem strength doesn't give the desired result. As far as associating a cable modem with the node is concerned, that idea has already been considered but modem might go down due to other reasons as well. You're right in that it can indicate a problem but it doesn't give the exact required output.

Log in or register to post comments