Output discards on ISL ports on MetroCluster IP back-end switches
Applies to
- ONTAP 9
- MetroCluster IP
- Back-end switches
Issue
- Back-end switches report increasing number of output discards on the ISL link(s):
CiscoA1# show interface Ethernet 1/19
Ethernet1/19 is up
..
TX
72364381398 unicast packets 246446 multicast packets 1439 broadcast packets
72364629283 output packets 228470689444346 bytes
27435311682 jumbo packets
0 output error 0 collision 0 deferred 0 late collision
0 lost carrier 0 no carrier 0 babble 7905858 output discard
0 Tx pause- Ports, connected to the controllers, are sending pause frames to slow down the incoming traffic:
CiscoA1# show interface Ethernet 1/9
Ethernet1/9 is up
..
TX
22205893584 unicast packets 246446 multicast packets 2558 broadcast packets
22206142588 output packets 89170173974575 bytes
10609662061 jumbo packets
0 output error 0 collision 0 deferred 0 late collision
0 lost carrier 0 no carrier 0 babble 0 output discard
4687218 Tx pause
- Transient
takeover disabledreported due to unsynchronized NVLOG:
Wed Jul 02 05:05:01 +0200 [nodeA2: cf_main: cf.fsm.takeoverOfPartnerDisabled:error]: Failover monitor: takeover of nodeA1 disabled (unsynchronized log).
- End-users might experience latency and/or slowness