Jumbo frame & pve-kernel-4.15.17-3-pve not work

[taf_321]

I'm using the PVE 5.2 test cluster on HP Proliant DL-160G6 servers. I recently updated the kernel to one of the kernel nodes before the version of pve-kernel-4.15.17-3-pve and I had problems with the network.

Network cards:

05:00.0 Ethernet controller: Intel Corporation 82576 Gigabit Network Connection (rev 01)
        Subsystem: Hewlett-Packard Company NC362i Integrated Dual port Gigabit Server Adapter
        Flags: bus master, fast devsel, latency 0, IRQ 24
        Memory at c0400000 (32-bit, non-prefetchable) [size=128K]
        Memory at c0420000 (32-bit, non-prefetchable) [size=128K]
        I/O ports at e880
        Memory at c04c0000 (32-bit, non-prefetchable) [size=16K]
        Expansion ROM at c0440000 [disabled] [size=128K]
        Capabilities: [40] Power Management version 3
        Capabilities: [50] MSI: Enable- Count=1/1 Maskable+ 64bit+
        Capabilities: [70] MSI-X: Enable+ Count=10 Masked-
        Capabilities: [a0] Express Endpoint, MSI 00
        Capabilities: [100] Advanced Error Reporting
        Capabilities: [140] Device Serial Number 1c-c1-de-ff-ff-08-59-b0
        Capabilities: [150] Alternative Routing-ID Interpretation (ARI)
        Capabilities: [160] Single Root I/O Virtualization (SR-IOV)
        Kernel driver in use: igb
        Kernel modules: igb

05:00.1 Ethernet controller: Intel Corporation 82576 Gigabit Network Connection (rev 01)
        Subsystem: Hewlett-Packard Company NC362i Integrated Dual port Gigabit Server Adapter
        Flags: bus master, fast devsel, latency 0, IRQ 35
        Memory at c0460000 (32-bit, non-prefetchable) [size=128K]
        Memory at c0480000 (32-bit, non-prefetchable) [size=128K]
        I/O ports at ec00
        Memory at c0504000 (32-bit, non-prefetchable) [size=16K]
        Expansion ROM at c04a0000 [disabled] [size=128K]
        Capabilities: [40] Power Management version 3
        Capabilities: [50] MSI: Enable- Count=1/1 Maskable+ 64bit+
        Capabilities: [70] MSI-X: Enable+ Count=10 Masked-
        Capabilities: [a0] Express Endpoint, MSI 00
        Capabilities: [100] Advanced Error Reporting
        Capabilities: [140] Device Serial Number 1c-c1-de-ff-ff-08-59-b0
        Capabilities: [150] Alternative Routing-ID Interpretation (ARI)
        Capabilities: [160] Single Root I/O Virtualization (SR-IOV)
        Kernel driver in use: igb
        Kernel modules: igb

Network settings:

auto lo
iface lo inet loopback

iface enp5s0f0 inet manual

auto enp5s0f1
iface enp5s0f1 inet static
        address  10.0.100.1
        netmask  255.255.255.0
        post-up ifconfig enp5s0f1 mtu 9000

auto vmbr0
iface vmbr0 inet static
        address 172.16.1.2
        netmask 255.255.255.0
        gateway 172.16.1.1
        bridge_ports enp5s0f0
        bridge_stp off
        bridge_fd 0

The interface enp5s0f1 looks towards the cluster network ceph and for the performance increase the transmission of larger frames is enabled.

After updating the kernel to pve-kernel-4.15.17-3-pve_4.15.17-12 on the interface, the RX errors counter began to grow and when executing ethtool -S enp5s0f1 I see the error growth at rx_length_errors. And also when executing

# pvesm status
got timeout
Name         Type     Status           Total            Used       Available        %
local         dir     active        28510260         2208896        24830084    7.75%
rdb           rbd     inactive            0               0              0        0%

And also when doing. In the log files, persistent messages

Jun 18 06:27:02 hekata pvestatd[1469]: got timeout
Jun 18 06:27:02 hekata pvestatd[1469]: status update time (5.069 seconds)
Jun 18 06:27:12 hekata pvestatd[1469]: got timeout
Jun 18 06:27:12 hekata pvestatd[1469]: status update time (5.067 seconds)
Jun 18 06:27:22 hekata pvestatd[1469]: got timeout
Jun 18 06:27:22 hekata pvestatd[1469]: status update time (5.068 seconds)
Jun 18 06:27:33 hekata pvestatd[1469]: got timeout
Jun 18 06:27:33 hekata pvestatd[1469]: status update time (5.068 seconds)
Jun 18 06:27:42 hekata pvestatd[1469]: got timeout
Jun 18 06:27:42 hekata pvestatd[1469]: status update time (5.067 seconds)
Jun 18 06:27:52 hekata pvestatd[1469]: got timeout
Jun 18 06:27:52 hekata pvestatd[1469]: status update time (5.067 seconds)

Return back to version pve-kernel-4.15.17-1-pve_4.15.17-9 returns the system health.

 

[Alwin]

Please put your output in [ QUOTE]output[ /QUOTE] or [ CODE]output[ /CODE] for better readability. You can find them under the + in the edit bar.

See here for more information on the intel drivers.
https://forum.proxmox.com/threads/4...or-pve-5-x-available.42097/page-6#post-212557