network performance issue appears periodically on PVE host

[RomanV]

Hello everyone.

Couple times in a month I got my network adapter works with ~7Mb/s speed.

ifconfig eth1 down && ifconfig eth1 up

temporary solves the problem for me.

I added iperf speed monitoring to zabbix and it looks like this:


I want find the reason of a that kind behaviour.
There is part of syslog just before it happend. Maybe it can help.

Apr  9 18:00:48 proxmox kernel: irq 16: nobody cared (try booting with the "irqpoll" option)
Apr  9 18:00:48 proxmox kernel: Pid: 0, comm: swapper veid: 0 Not tainted 2.6.32-18-pve #1
Apr  9 18:00:48 proxmox kernel: Call Trace:
Apr  9 18:00:48 proxmox kernel: <IRQ>  [<ffffffff810ecbab>] ? __report_bad_irq+0x2b/0xa0
Apr  9 18:00:48 proxmox kernel: [<ffffffff810ecdac>] ? note_interrupt+0x18c/0x1d0
Apr  9 18:00:48 proxmox kernel: [<ffffffff810ed4bd>] ? handle_fasteoi_irq+0xcd/0xf0
Apr  9 18:00:48 proxmox kernel: [<ffffffff8100df79>] ? handle_irq+0x49/0xa0
Apr  9 18:00:48 proxmox kernel: [<ffffffff8152492c>] ? do_IRQ+0x6c/0xf0
Apr  9 18:00:48 proxmox kernel: [<ffffffff8100ba93>] ? ret_from_intr+0x0/0x11
Apr  9 18:00:48 proxmox kernel: <EOI>  [<ffffffff812d0e7e>] ? intel_idle+0xde/0x170
Apr  9 18:00:48 proxmox kernel: [<ffffffff812d0e61>] ? intel_idle+0xc1/0x170
Apr  9 18:00:48 proxmox kernel: [<ffffffff81420387>] ? cpuidle_idle_call+0xa7/0x140
Apr  9 18:00:48 proxmox kernel: [<ffffffff8100a023>] ? cpu_idle+0xb3/0x110
Apr  9 18:00:48 proxmox kernel: [<ffffffff81503e35>] ? rest_init+0x85/0x90
Apr  9 18:00:48 proxmox kernel: [<ffffffff81c2ef6e>] ? start_kernel+0x412/0x41e
Apr  9 18:00:48 proxmox kernel: [<ffffffff81c2e33a>] ? x86_64_start_reservations+0x125/0x129
Apr  9 18:00:48 proxmox kernel: [<ffffffff81c2e438>] ? x86_64_start_kernel+0xfa/0x109
Apr  9 18:00:48 proxmox kernel: handlers:
Apr  9 18:00:48 proxmox kernel: [<ffffffffa0028190>] (rtl8169_interrupt+0x0/0x3e0 [r8169])
Apr  9 18:00:48 proxmox kernel: Disabling IRQ #16

# pveversion -v
pve-manager: 2.3-13 (pve-manager/2.3/7946f1f1)
running kernel: 2.6.32-18-pve
proxmox-ve-2.6.32: 2.3-88
pve-kernel-2.6.32-11-pve: 2.6.32-66
pve-kernel-2.6.32-18-pve: 2.6.32-88
lvm2: 2.02.95-1pve2
clvm: 2.02.95-1pve2
corosync-pve: 1.4.4-4
openais-pve: 1.1.4-2
libqb: 0.10.1-2
redhat-cluster-pve: 3.1.93-2
resource-agents-pve: 3.9.2-3
fence-agents-pve: 3.1.9-1
pve-cluster: 1.0-36
qemu-server: 2.3-18
pve-firmware: 1.0-21
libpve-common-perl: 1.0-48
libpve-access-control: 1.0-26
libpve-storage-perl: 2.3-6
vncterm: 1.0-3
vzctl: 4.0-1pve2
vzprocps: 2.0.11-2
vzquota: 3.1-1
pve-qemu-kvm: 1.4-8
ksm-control-daemon: 1.1-1

Any help is greatly appreciated.

 

[booleanxor]

Same issue here, same error. Have you found anything?

My post here: https://forum.proxmox.com/threads/slow-networking-until-pve-reboot.91513/

Just tried:

GRUB_CMDLINE_LINUX_DEFAULT="quiet intel_iommu=on irqpoll"

It happens at EXACTLY the time that my network performance goes down the tubes.

 

[booleanxor]

I tracked it down to a bad realtek driver (seems like we have the same one). Tried getting a new one from realtek and compiling but apparently there's just issues with it. I'm just going to get a new intel nic. Read my thread for more details.