rx_brb_discards on QLogic 3442-CU-CK

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

Stril

Member
Sep 26, 2017
191
12
18
41

Hi!

I did an upgrade of two linux-systems. Since the update, I have exactly the same problem on both systems:

The first port of the cards shows:

rx_brb_discard: 14
rx_brb_truncate: 2


--> rx-Errors


The systems to not have a lot of load.
As its 'eth3' that is affected in both systems and both are connected to different switches, I can eliminate a hardware-problem.

Did you ever see that behaviour?
Do you have any idea on how to solve this?

Thank you for your help!

Stril

 

altmind

Active Member
Sep 23, 2018
285
101
43
Environment
  • Red Hat Enterprise Linux 6
Issue
  • See small, but persistent increases in RX errors on multi-core systems with 10 Gbit NICs
  • RX frame errors on bnx2x interfaces
Resolution
  • If physical network bottlenecks are not suspected, disabling CPU C-States in both the BIOS and the kernel may alleviate these errors.
  • For more information on C-States, please see What are CPU "C-states" and how to disable them if needed?
    To limit a CPU to a certain C-state, you can pass the processor.max_cstate=X option in the kernel line of /boot/grub/grub.conf in RHEL 6 and below, or in /etc/sysconfig/grub, on the GRUB_CMDLINE_LINUX line in RHEL 7.
Root Cause
It is possible for network hardware buffers to fill and overflow while under extreme network load as power management wakes up idle CPUs. This appears most likely to occur on multi-core systems with relatively low CPU load and very high network load.

This issue has been seen and resolved on Dell PowerEdge R9xx and R8xx with bnx2x NICs.