snmp port status X670-48x

[AntonR]

how can I read the status of the ports via SNMP. oid .1.3.6.1.4.1.1916.1.4.2.1.3 not working.

snmpwalk -v 2c -c a7V2Y9W 10.147.220.250 .1.3.6.1.4.1.1916.1.4.2.1.3
SNMPv2-SMI::enterprises.1916.1.4.2.1.3 = No Such Object available on this agent at this OID

ExtremeXOS version 16.2.4.5 by release-manager

[EtherMAN]

It is surprising on how much polling even the lower performance models like the 430 can take before it impacts anything. We have Cacti pulling port stats for bandwidth. errors. temp. Solarwinds for sflow. SNMPC for traps and switch to switch links. Extreme NMS for configs and management . All grabbing stuff every 5 minutes on about 12K interfaces 3k or so devices. Everything is still happy and doing it's job LOL.

[jeronimo]

Well, there's the concept of snmptable which allows you to get all the data, make easy correlation between the fields, and finally extract what you need.

code:

# snmptable switch IF-MIB::ifTable
SNMP table: IF-MIB::ifTable

 ifIndex                                 ifDescr         ifType ifMtu    ifSpeed  ifPhysAddress ifAdminStatus ifOperStatus  ifLastChange ifInOctets ifInUcastPkts ifInNUcastPkts ifInDiscards ifInErrors ifInUnknownProtos ifOutOctets ifOutUcastPkts ifOutNUcastPkts ifOutDiscards ifOutErrors ifOutQLen           ifSpecific
    1001                   X690-48x-2q-4c Port 1 ethernetCsmacd  1500          0 0:4:96:af:51:2          down         down  0:0:00:24.00          0             0              0            0          0                 0           0              0               0             0           0         0 SNMPv2-SMI::internet

...

Of course, polling the entire table each time is quite heavy on the SNMP engine, so snmpget'ing specific values might be better. It all depends on the implementation (caching etc.) Each time you are confronted with these tasks it feels like a dark art ?

[EtherMAN]

We use snpmc and use that to get link status on all trunk links between switches.. This gives our NOC more intel about the link other than just a port showing down. Here is how we have it set in our polling server. I don't know what you are doing your mib walks and polling with but the trick is to get the right interface id (1001) for port 1 and so on ... and set normal status to up to get a trap with link specific data you have labeled in the link object. We always Poll closest switch to core. Not sure what other attributes you may be trying to get from the port.

Still can't poll light levels via a mib form any of the optical links or ports like many other vendors can. Would be great to be able to graph those levels for desegregation.

[jeronimo]

Why not simply use IF-MIB::ifOperStatus:

.1.3.6.1.2.1.2.2.1.8.1001 = INTEGER: down(2)
.1.3.6.1.2.1.2.2.1.8.1002 = INTEGER: down(2)
.1.3.6.1.2.1.2.2.1.8.1003 = INTEGER: down(2)
.1.3.6.1.2.1.2.2.1.8.1004 = INTEGER: down(2)