Error disabled. Reason:DCX-No ACK in 100 PDUs

[thomas_daley]

Hi,

  I have a customer who lost all connectivity from the ESX host for both networking and FCoE because (as the title suggests) the interfaces were error disabled.  This happened across all 8, dual ported, dual homed CNAs at the same time.  Does anyone have any idea what this error comes from?  The are using ESX 4.0 and are running Nexus 5020 with 4.2(1)N2(1a).

Thanks,

Thom

[Prashanth Krishnappa]

Hello Evan

This issue in my experience is almost always a host issue. More than "Ack no: 0" being an issue its about the host responding with same Seq and Ack number(00000001 00000000)

Thanks

-Prashanth

[evanweston]

Thanks Prasanth,

Contacted VMWare and they picked up that the Emulex OCe10100 CNA is not supported in ESX.

Changed to a supported CNA and we've been good for over 2 hours now.

Evan

[mitra dray]

i wonder what the behavior of disabling the dcbx tlv on the server side would be. 
as i undeesrand it from reading linux man 
i see that there are 2 types of dcbx legacy and new 
a kind of fallback between the two talking to the switch. if the switch knows only "new" it could also be an explanation (in addition to simple timeout).

CEE - CEE DCBX will be used if lldpad is restarted

CIN - CIN DCBX will be used if lldpad is restarted

we can try 

Disable transmission of the CEE DCBX TLV
lldptool -T -i eth3 -V CEE-DCBX enableTx=no

Disable transmission of the CIN DCBX TLV
lldptool -T -i eth3 -V CIN-DCBX enableTx=no

i wonder.