Ethdev port_id=0 nb_tx_queues=10 > 8 EAL: Error - exiting with code: 1

[Mark Wittling]

I am trying to run Trex in an ESXi Virtual Machine (Cent7), with as many cores as possible. The VM has as many as 32 cores on it.

I can run with 4 cores, but that seems to be the limit. As I try to move up past 4 cores, I either get a "power of 2" error (ERROR: Number of rx queues not power of 2) if the cores are, say, 5, 6 or 7. But if you set cores set to 8, I get a different unknown EAL error.

# ./t-rex-64 -i --stl
Starting Scapy server.... Scapy server is started
The ports are bound/configured.
Starting  TRex v2.97 please wait  ...
 set driver name net_vmxnet3
 driver capability  : TSO  LRO
 set dpdk queues mode to MULTI_QUE
 Number of ports found: 2
zmq publisher at: tcp://*:4500
Ethdev port_id=0 nb_tx_queues=10 > 8
EAL: Error - exiting with code: 1
  Cause: Cannot configure device: err=-22, port=0
Killing Scapy server... Scapy server is killed

Can VMWare not do more than 4 cores with Trex?

[hanoh haim]

There is a limit to the amount of queues per driver, try to add more virtual interfaces 

--
You received this message because you are subscribed to the Google Groups "TRex Traffic Generator" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trex-tgn+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/trex-tgn/279c8348-9590-447e-9614-670a32da521dn%40googlegroups.com.

--

Hanoh

Sent from my iPhone

[Mark Wittling]

So this is a driver limitation? For vmxnet3?

There's no way to ramp up the queues for the driver? Or, switch to a driver that can handle more queues?

[hanoh haim]

Hi Mark, 

I'm not familiar with VMXNET3 but maybe you can add more queues in the hypervisor and the driver just read this configuration 

Thanks

Hanoh