Regression/bug on kernel 4.13.13-6-pve

[mgiammarco]

Hello,
I have a dell r710 server with ibm mellanox pcie 10g nic:

07:00.0 Ethernet controller: Mellanox Technologies MT26448 [ConnectX EN 10GigE, PCIe 2.0 5GT/s] (rev a0)
Subsystem: Mellanox Technologies MT26448 [ConnectX EN 10GigE, PCIe 2.0 5GT/s]
Flags: bus master, fast devsel, latency 0, IRQ 37
Memory at df300000 (64-bit, non-prefetchable) [size=1M]
Memory at d5000000 (64-bit, prefetchable) [size=8M]
Expansion ROM at df200000 [disabled] [size=1M]
Capabilities: [40] Power Management version 3
Capabilities: [48] Vital Product Data
Capabilities: [9c] MSI-X: Enable+ Count=256 Masked-
Capabilities: [60] Express Endpoint, MSI 00
Capabilities: [100] Alternative Routing-ID Interpretation (ARI)
Kernel driver in use: mlx4_core
Kernel modules: mlx4_core


Until kernel 4.13.13-5-pve it works perfectly.
From kernel 4.13.13-6-pve it stops working, see dmesg:

Feb 19 06:25:13 bigtwo kernel: [54552.495520] mlx4_en: enp7s0d1: Link Up
Feb 19 06:25:13 bigtwo kernel: [54552.600407] mlx4_en: enp7s0d1: Link Down
Feb 19 06:25:13 bigtwo kernel: [54552.655406] mlx4_en: enp7s0d1: Link Up
Feb 19 06:25:13 bigtwo kernel: [54552.710348] mlx4_en: enp7s0d1: Link Down
Feb 19 06:25:13 bigtwo kernel: [54552.726686] vmbr1: port 2(enp7s0d1) entered disabled state
Feb 19 06:25:14 bigtwo kernel: [54552.815317] mlx4_en: enp7s0d1: Link Up
Feb 19 06:25:14 bigtwo kernel: [54552.823501] vmbr1: port 2(enp7s0d1) entered blocking state
Feb 19 06:25:14 bigtwo kernel: [54552.823505] vmbr1: port 2(enp7s0d1) entered listening state
Feb 19 06:25:14 bigtwo kernel: [54553.230060] mlx4_en: enp7s0d1: Link Down
Feb 19 06:25:14 bigtwo kernel: [54553.485010] mlx4_en: enp7s0d1: Link Up
Feb 19 06:25:14 bigtwo kernel: [54553.644909] mlx4_en: enp7s0d1: Link Down
Feb 19 06:25:14 bigtwo kernel: [54553.750592] vmbr1: port 2(enp7s0d1) entered disabled state
Feb 19 06:25:15 bigtwo kernel: [54553.854769] mlx4_en: enp7s0d1: Link Up
Feb 19 06:25:15 bigtwo kernel: [54553.863005] vmbr1: port 2(enp7s0d1) entered blocking state
Feb 19 06:25:15 bigtwo kernel: [54553.863010] vmbr1: port 2(enp7s0d1) entered listening state
Feb 19 06:25:15 bigtwo kernel: [54554.019739] mlx4_en: enp7s0d1: Link Down
Feb 19 06:25:15 bigtwo kernel: [54554.384586] mlx4_en: enp7s0d1: Link Up
Feb 19 06:25:15 bigtwo kernel: [54554.439533] mlx4_en: enp7s0d1: Link Down
Feb 19 06:25:15 bigtwo kernel: [54554.494478] mlx4_en: enp7s0d1: Link Up
Feb 19 06:25:16 bigtwo kernel: [54554.819297] mlx4_en: enp7s0d1: Link Down


I have other servers with other mellanox cards (hp branded) and they works.

Mario

 

[Alwin]

Are you using the newest FW on the card? Any special, network or sysctl settings?

 

[mgiammarco]

I have firmware 1.5.5 I do not know how I can upgrade.
I have no special network settings nor sysctl.