Increasing number of inbound discards on new APA failover group

[chouse]

Recently created an APA failovergroup with lan900 using lan0 and lan3 as uplinks:

# nwmgr -S apa -I 900 -v
lan900 current values:
Speed = 1 Gbps Full Duplex
MTU = 1500
Virtual Maximum Transmission Unit = 0
MAC Address = 0x001e0b5c2500
Network Management ID = 6
Features = Linkagg Interface
IPV4 Recv CKO
IPV4 Send CKO
VLAN Support
VLAN Tag Offload
64Bit MIB Support
Load Distribution Algorithm = LB_HS
Mode = LAN_MONITOR
Parent PPA = -
APA State = Up
Membership = 0,3
Active Port(s) = 0
Ready Port(s) = 3
Not Ready Port(s) = -
Connected Port(s) = 3
Polling Interval = 1000000
Dead Count = 2
Rapid ARP = on
Rapid ARP Interval = 1.0 second(s)
Rapid ARP Count = 10
Failover Policy = PRIORITY_BASED

Active port is lan0, ready port is lan3.

Seeing a lot of inbound discards:

# nwmgr --st -I 900 -S apa

*** lan900 64 bit MIB statistics:
Interface Name = lan900
PPA Number = 900
Description = lan900 Hewlett-Packard LinkAggregate Interface
MTU Size = 1500
Speed = 1 Gbps
Station Address = 0x001E0B5C2500
Administration Status = UP
Operation Status = UP
Last Change = Tue Feb 7 13:29:35 2012
Inbound Octets = 973020617
Inbound Unicast Packets = 4843469
Inbound Multicast Packets = 0
Inbound Broadcast Packets = 3057991
Inbound Discards = 3058260
Inbound Errors = 0
Inbound Unknown Protocols = 109141
Outbound Octets = 7439997121
Outbound Unicast Packets = 7359557
Outbound Multicast Packets = 0
Outbound Broadcast Packets = 269
Outbound Discards = 0
Outbound Errors = 0
Counter Discontinuity Time = Sat Nov 12 18:41:25 2011
Physical Promiscuous Mode = FALSE
Physical Connector Present = FALSE
Interface Alias =
Link Up/Down Trap Enable = Enabled

Discards are visible on the non-active uplink only (lan3):

# snmpwalk -v 1 -c public 10.142.10.16 .1.3.6.1.2.1.2.2.1.2 | head -n 6
iso.3.6.1.2.1.2.2.1.2.1 = STRING: "lan2 HP PCI-X 1000Mbps Release B.11.31.0903"
iso.3.6.1.2.1.2.2.1.2.2 = STRING: "lan0 HP PCI-X 1000Mbps Release B.11.31.0903"
iso.3.6.1.2.1.2.2.1.2.3 = STRING: "lan1 HP PCI-X 1000Mbps Release B.11.31.0903"
iso.3.6.1.2.1.2.2.1.2.4 = STRING: "lan3 HP PCI-X 1000Mbps Release B.11.31.0903"
iso.3.6.1.2.1.2.2.1.2.5 = STRING: "software loopback"
iso.3.6.1.2.1.2.2.1.2.6 = STRING: "lan900 Hewlett-Packard LinkAggregate Interface"

# snmpwalk -v 1 -c public 10.142.10.16 .1.3.6.1.2.1.2.2.1.13 | head -n 6
iso.3.6.1.2.1.2.2.1.13.1 = Counter32: 0
iso.3.6.1.2.1.2.2.1.13.2 = Counter32: 0
iso.3.6.1.2.1.2.2.1.13.3 = Counter32: 0
iso.3.6.1.2.1.2.2.1.13.4 = Counter32: 3090141
iso.3.6.1.2.1.2.2.1.13.5 = Counter32: 0
iso.3.6.1.2.1.2.2.1.13.6 = Counter32: 3061053

Is this normal/expected behavior ?

[chouse · Accepted solution]

To answer my own question, I opened a case with HP and confirmed that seeing these inbound discards is normal behavior:

"Inbound discards on the LAN Monitor standby link or on the LAN Monitor Failover Group do not indicate an issue. It is normal to see them.

The standby port receives broadcast packets and multicast packets exactly the same as the ones received by the Active port. However, here is no IP protocol bound or running on the Standby link. The fact is that we don’t want to handle the same packet twice. This implies that the standby port will discard all multicast and broadcast packets. The standby port will also discard all Unicast packets that might somehow be sent to it."