I increased RX ring params on the interface and restarted networking on each host.  So far the error counts on all three hosts 1gig interfaces are still at zero.  Will see how it holds up

On Thu, Jun 6, 2019 at 12:20 PM Jayme <jaymef@gmail.com> wrote:
I have a three node HCI setup on Dell R720s running the latest stable version of 4.3.3

Each hosts has a 1gig link and a 10gig link.  The 1gig is used for ovirt management network and 10gig link is used for backend glusterFS traffic.

I haven't noticed before but after installing ovirt metrics store I'm seeing that gig interface used for ovirtmgmt on all three hosts are showing high RX error rates.  The 10gig interfaces for glusterFS on all three hosts appear to be fine. 

The 1gig ethernet controllers are: Broadcom Inc. and subsidiaries NetXtreme II BCM57800 1/10 Gigabit Ethernet (rev 10)

Other physical servers on the same network/switches outside of oVirt have zero RX errors. 

Here is an example of what I'm seeing:

host0:

# ip -s link show em3
4: em3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq master ovirtmgmt state UP mode DEFAULT group default qlen 1000
    link/ether b0:83:fe:cc:9a:2d brd ff:ff:ff:ff:ff:ff
    RX: bytes  packets  errors  dropped overrun mcast
    51777532544474 36233202312 416993  0       0       2062421
    TX: bytes  packets  errors  dropped carrier collsns
    7284362442704 18685883330 0       0       0       0

host1:

# ip -s link show em3
4: em3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq master ovirtmgmt state UP mode DEFAULT group default qlen 1000
    link/ether b0:83:fe:cc:99:31 brd ff:ff:ff:ff:ff:ff
    RX: bytes  packets  errors  dropped overrun mcast
    9518766859330 14424644226 89638   0       0       2056578
    TX: bytes  packets  errors  dropped carrier collsns
    27866585257227 22323979969 0       0       0       0

host2:

# ip -s link show em3
4: em3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq master ovirtmgmt state UP mode DEFAULT group default qlen 1000
    link/ether b0:83:fe:cc:92:50 brd ff:ff:ff:ff:ff:ff
    RX: bytes  packets  errors  dropped overrun mcast
    6409138012195 13045254148 14825   0       0       2040655
    TX: bytes  packets  errors  dropped carrier collsns
    31577745516683 23466818659 0       0       0       0

Anyone have any ideas why the RX error rate on the ovirtmgmt network could be so high?