I'm doing a product demo for one of our customers.
They've setup RHEV 3.5 on a Dell Blade center. We also used 3.4 to
start this demo. So far they've been less than impressed as we have 10
VM's that are just putting some traffic on the system. Under any load
the blades network card driver fails and the VMs are then paused.
Under ESX the same blades have no issues at all. We started with Dell
on this but were unable to find any issues with these systems related
to hardware.
Running on the 3.5 node (even setup one blade as a RHEL + VDSM) and we
continue to get the same errors.
They continue to get this on the 3.5 node image, 3.4 node image,
RHEL{6..7}+VDSM, CentOS{6..7}
Apr 7 23:09:15 POCserver2 kernel: qlcnic 0000:01:00.1: Pause control
frames disabled on all ports
Apr 7 23:09:15 POCserver2 kernel: qlcnic 0000:01:00.1: firmware hang detected
Apr 7 23:09:15 POCserver2 kernel: qlcnic 0000:01:00.1: Dumping hw/fw registers
Apr 7 23:09:15 POCserver2 kernel: PEG_HALT_STATUS1: 0x40001502,
PEG_HALT_STATUS2: 0x3dd980,
Apr 7 23:09:15 POCserver2 kernel: PEG_NET_0_PC: 0x6d394, PEG_NET_1_PC: 0x6d466,
Apr 7 23:09:15 POCserver2 kernel: PEG_NET_2_PC: 0x149, PEG_NET_3_PC: 0x6e598,
Apr 7 23:09:15 POCserver2 kernel: PEG_NET_4_PC: 0x12268
Apr 7 23:09:15 POCserver2 kernel: qlcnic 0000:01:00.0: Pause control
frames disabled on all ports
Apr 7 23:09:15 POCserver2 kernel: qlcnic 0000:01:00.0: firmware hang detected
Apr 7 23:09:15 POCserver2 kernel: qlcnic 0000:01:00.0: Dumping hw/fw registers
Apr 7 23:09:15 POCserver2 kernel: PEG_HALT_STATUS1: 0x40001502,
PEG_HALT_STATUS2: 0x3dd980,
Apr 7 23:09:15 POCserver2 kernel: PEG_NET_0_PC: 0x6d394, PEG_NET_1_PC: 0x6d466,
Apr 7 23:09:15 POCserver2 kernel: PEG_NET_2_PC: 0x149, PEG_NET_3_PC: 0x6e598,
Apr 7 23:09:15 POCserver2 kernel: PEG_NET_4_PC: 0x12268
I've really got limited time for this POC so I sent a support case
with Red Hat as well. Just hoped that this community may have seen
this. Thus far my Googlefoo has failed me.