My current setting is:
Host: CentOS 6.4
Mellanox Driver: MLNX_OFED_LINUX-2.0-3.0.0-rhel6.4-x86_64 with SRIOV enabled
Virtualization: KVM
Mellanox Card: ConnectX-3 EN
So far so good for last 2 months, I haven't seen any error messages about Mellanox printed to the screen directly. Yesterday, I saw the following messages printed on the screen as well as to the syslog(/var/log/messages).
localhost kernel: mlx4_core 0000:09:00.0: command ACCESS_MEM (0x2e) failed: in_param=0x208b36001, in_mod=0x100, op_mod=0x0, fw status = 0x1
localhost kernel: mlx4_core 0000:09:00.0: mlx4_master_process_vhcr:Failed reading vhcrret: 0xfffffffb
localhost kernel: mlx4_core 0000:09:00.0: Failed processing vhcr for slave: 1, resetting slave
If any one could please let me know if the above error message is a bug in the driver? or it just a warning?
Help is appreciated. Thanks!