OVS Warnung nach Update auf 5.2

pixelpeter

Well-Known Member
Aug 5, 2015
168
4
58
57
Chemnitz
Hallo,


Seit dem Update auf 5.2 bekomme ich immer beim starten einer VM diese Meldung:

[ 2319.264555] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.

Kann man das ignorieren?


Peter
 
Welcher Kernel und welche Openvswitch Version ist den installiert?
 
Is there anything more in the logs?
 
Nein.

[ 155.946099] device tap100i0 entered promiscuous mode
[ 155.963465] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
[ 160.440843] device tap103i0 entered promiscuous mode
[ 160.455800] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
[ 162.503676] usb 3-1: USB disconnect, device number 2
[ 165.456647] device tap113i0 entered promiscuous mode
[ 165.472018] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
[ 165.848224] device tap113i1 entered promiscuous mode
[ 165.864139] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
[ 170.498246] device tap119i0 entered promiscuous mode
[ 170.513666] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
[ 174.534383] device tap122i0 entered promiscuous mode
[ 174.549513] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
[ 179.513559] device tap123i0 entered promiscuous mode
[ 179.559422] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
[ 184.534106] device tap134i0 entered promiscuous mode
[ 184.549848] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
[ 184.954796] device tap134i1 entered promiscuous mode
[ 184.970658] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
[ 189.542329] device tap151i0 entered promiscuous mode
[ 189.560729] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
[ 194.589510] device tap154i0 entered promiscuous mode
[ 194.605279] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
[ 199.544475] device tap155i0 entered promiscuous mode
[ 199.560160] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
[ 204.573447] device tap158i0 entered promiscuous mode
[ 204.589823] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
[ 204.994107] device tap158i1 entered promiscuous mode
[ 205.010826] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
[ 2308.223618] device tap102i0 entered promiscuous mode
[ 2308.241500] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
[ 2309.839396] device tap104i0 entered promiscuous mode
[ 2309.856268] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
[ 2312.143422] device tap101i0 entered promiscuous mode
[ 2312.160929] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
[ 2319.247637] device tap142i0 entered promiscuous mode
[ 2319.264555] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.
[ 4726.090295] perf: interrupt took too long (2509 > 2500), lowering kernel.perf_event_max_sample_rate to 79500
[ 7685.045711] device tap118i0 entered promiscuous mode
[ 7685.064130] netlink: 'ovs-vswitchd': attribute type 5 has an invalid length.


Peter
 
Ist das openvswitch module geladen?
 
Bitte poste die Netzwerk und OVS spezifische Config.
 
Hier die Netwerkkonfig.
OVS spezifisch haben wir nichts konfiguriert.

# network interface settings; autogenerated
# Please do NOT modify this file directly, unless you know what
# you're doing.
#
# If you want to manage part of the network configuration manually,
# please utilize the 'source' or 'source-directory' directives to do
# so.
# PVE will preserve these directives, but will NOT its network
# configuration from sourced files, so do not attempt to move any of
# the PVE managed interfaces into external files!

source /etc/network/interfaces.d/*

auto lo
iface lo inet loopback

allow-hotplug eth0
allow-hotplug eth1
allow-hotplug eth2
allow-hotplug eth3
allow-hotplug eth4
allow-hotplug eth5

iface eth0 inet manual
iface eth1 inet manual
iface eth2 inet manual
iface eth3 inet manual
iface eth4 inet manual
iface eth5 inet manual

# Cluster Kommunikation
auto bond0
iface bond0 inet static
address 10.6.9.211
netmask 255.255.255.0
gateway 10.6.9.2
slaves eth0 eth1
bond_miimon 100
bond_mode active-backup

# Storage
auto bond1
iface bond1 inet static
address 10.3.9.211
netmask 255.255.255.0
slaves eth2 eth3
bond_miimon 100
bond_mode active-backup

# VM
allow-vmbr0 bond2
iface bond2 inet manual
ovs_bonds eth4 eth5
ovs_type OVSBond
ovs_bridge vmbr0
ovs_options bond_mode=active-backup

auto vmbr0
iface vmbr0 inet manual
ovs_type OVSBridge
ovs_ports bond2


Peter
 
Ich kann die Meldung auf meinem System leider nicht nachvollziehen. Ist im Log etwas zu sehen? -> /var/log/openvswitch/ovs-vswitchd.log

Vielleicht hilft das aufdrehen des Logging.
http://ask.xmodulo.com/enable-logging-open-vswitch.html

# VM
allow-vmbr0 bond2
iface bond2 inet manual
ovs_bonds eth4 eth5
ovs_type OVSBond
ovs_bridge vmbr0
ovs_options bond_mode=active-backup

auto vmbr0
iface vmbr0 inet manual
ovs_type OVSBridge
ovs_ports bond2
Wenn openvswitch nicht für mehr gebraucht wird, dann könnte dieser Teil auch mit der Linux Bridge ersetzt werden (kann auch VLAN).

EDIT: Der Inhalt der OVS Datenbank kann mit folgendem Befehl abgerufen werden 'ovs-vsctl show'.
 
Last edited:
Hallo Alwin,

Bei unser letzten Proxmox Schulung war die Aussage das Bridges sich wie Hubs verhalten und damit sämtlicher Traffic in den VM's "sichtbar" wäre.
Ist das noch so? Wenn nicht dann werden wir in Zukunft gerne auf OVS verzichten.

Hier der Log:

2018-05-23T04:25:01.102Z|00096|vlog|INFO|opened log file /var/log/openvswitch/ovs-vswitchd.log
2018-05-23T04:50:28.923Z|00126|ofproto_dpif_xlate(handler52)|WARN|received packet on unknown port 3 while processing icmp6,in_port=3,vlan_tci=0x0000,dl_src=e2:61:df:90:d7:7b,dl_dst=33:33:00:00:00:02,ipv6_src=fe80::e061:dfff:fe90:d77b,ipv6_dst=ff02::2,ipv6_label=0x00000,nw_tos=0,nw_ecn=0,nw_ttl=255,icmp_type=133,icmp_code=0,nd_target=::,nd_sll=00:00:00:00:00:00,nd_tll=00:00:00:00:00:00 on bridge vmbr0
2018-05-23T05:53:50.011Z|00127|ofproto_dpif_xlate(handler52)|WARN|received packet on unknown port 3 while processing icmp6,in_port=3,vlan_tci=0x0000,dl_src=e2:61:df:90:d7:7b,dl_dst=33:33:00:00:00:02,ipv6_src=fe80::e061:dfff:fe90:d77b,ipv6_dst=ff02::2,ipv6_label=0x00000,nw_tos=0,nw_ecn=0,nw_ttl=255,icmp_type=133,icmp_code=0,nd_target=::,nd_sll=00:00:00:00:00:00,nd_tll=00:00:00:00:00:00 on bridge vmbr0
2018-05-23T06:59:22.171Z|00128|ofproto_dpif_xlate(handler52)|WARN|received packet on unknown port 3 while processing icmp6,in_port=3,vlan_tci=0x0000,dl_src=e2:61:df:90:d7:7b,dl_dst=33:33:00:00:00:02,ipv6_src=fe80::e061:dfff:fe90:d77b,ipv6_dst=ff02::2,ipv6_label=0x00000,nw_tos=0,nw_ecn=0,nw_ttl=255,icmp_type=133,icmp_code=0,nd_target=::,nd_sll=00:00:00:00:00:00,nd_tll=00:00:00:00:00:00 on bridge vmbr0
2018-05-23T07:56:10.043Z|00129|ofproto_dpif_xlate(handler52)|WARN|received packet on unknown port 3 while processing icmp6,in_port=3,vlan_tci=0x0000,dl_src=e2:61:df:90:d7:7b,dl_dst=33:33:00:00:00:02,ipv6_src=fe80::e061:dfff:fe90:d77b,ipv6_dst=ff02::2,ipv6_label=0x00000,nw_tos=0,nw_ecn=0,nw_ttl=255,icmp_type=133,icmp_code=0,nd_target=::,nd_sll=00:00:00:00:00:00,nd_tll=00:00:00:00:00:00 on bridge vmbr0
2018-05-23T08:52:57.915Z|00130|ofproto_dpif_xlate(handler52)|WARN|received packet on unknown port 3 while processing icmp6,in_port=3,vlan_tci=0x0000,dl_src=e2:61:df:90:d7:7b,dl_dst=33:33:00:00:00:02,ipv6_src=fe80::e061:dfff:fe90:d77b,ipv6_dst=ff02::2,ipv6_label=0x00000,nw_tos=0,nw_ecn=0,nw_ttl=255,icmp_type=133,icmp_code=0,nd_target=::,nd_sll=00:00:00:00:00:00,nd_tll=00:00:00:00:00:00 on bridge vmbr0

Peter
 
Der Inhalt der OVS Datenbank kann mit folgendem Befehl abgerufen werden 'ovs-vsctl show'. Vielleicht ist ein Eintrag in der Datenbank, der nicht mehr der Kompatibel ist.

Bei unser letzten Proxmox Schulung war die Aussage das Bridges sich wie Hubs verhalten und damit sämtlicher Traffic in den VM's "sichtbar" wäre.
Ist das noch so? Wenn nicht dann werden wir in Zukunft gerne auf OVS verzichten.
Ja, die Bridge läuft im Promiscuous Mode und der Traffic wird an alle Ports weitergeleitet.
 
Hallo Alwin,

Hier der Output:

08c67c19-c230-4a31-86b6-c29a2dd9c2b1
Bridge "vmbr0"
Port "tap158i0"
tag: 310
Interface "tap158i0"
Port "tap158i1"
tag: 311
Interface "tap158i1"
Port "bond2"
Interface "eth4"
Interface "eth5"
Port "tap151i0"
Interface "tap151i0"
Port "tap113i1"
tag: 311
Interface "tap113i1"
Port "tap154i0"
tag: 368
Interface "tap154i0"
Port "vmbr0"
Interface "vmbr0"
type: internal
Port "tap155i0"
tag: 308
Interface "tap155i0"
Port "tap102i0"
tag: 305
Interface "tap102i0"
Port "tap122i0"
tag: 305
Interface "tap122i0"
Port "tap103i0"
tag: 350
Interface "tap103i0"
Port "tap134i0"
tag: 310
Interface "tap134i0"
Port "tap119i0"
tag: 350
Interface "tap119i0"
Port "tap123i0"
tag: 350
Interface "tap123i0"
Port "tap134i1"
tag: 311
Interface "tap134i1"
Port "tap142i0"
Interface "tap142i0"
Port "tap104i0"
tag: 305
Interface "tap104i0"
Port "tap101i0"
tag: 305
Interface "tap101i0"
Port "tap100i0"
tag: 308
Interface "tap100i0"
Port "tap113i0"
tag: 310
Interface "tap113i0"
ovs_version: "2.7.0"

Wir setzen in den nächsten Tagen einen neuen Cluster auf.
Wenn die Warnung dort nicht vorhanden ist dann werden wir den vorhandenen bei Gelegenheit neu machen.
Der hat schon einige Releasesprünge hinter sich. Nicht das hier noch irgendwelche Altlasten vorhanden sind.
Aktuell läuft ja alles.

Danke für die Unterstützung.

Peter
 
Code:
root@openvswitch:~# ovs-vsctl show
d1c1be27-22f9-4064-9e4d-3ed931e010b9
    Bridge "vmbr1"
        Port "vmbr1"
            Interface "vmbr1"
                type: internal
        Port "tap100i0"
            Interface "tap100i0"
        Port "ens19"
            Interface "ens19"
    ovs_version: "2.7.0"
Auszug aus meinem Testsystem, den 'tag:' Eintrag gibt es hier nicht. Vielleicht ist dieser noch von einer alten OVS Version, der nicht mehr kompatibel ist.
 
Der Tag kommt meiner Meinung nach von dem gesetzten VLAN in der VM Konfiguration.
Stimmt. Mein Testsystem hat eine frische DB. ;) Was mir zum Schluss noch aufgefallen ist, mit dem Befehl 'ovs-vsctl list Interface', scheint ein 'type:' auf.
 

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!