Join Log in
Score:1
avatar Server

IPv6 multipath routing with configurable weight

us flag
m.divya.mohan

With Linux kernel 4.14, when I add IPv6 multipath routes with weights (10 & 30), they are getting added with same weight '1'. IPv4 route does not have this issue.

I checked with exact same steps in a different machine with Linux kernel 4.19, where the weights displayed in output seems to match the configuration.

Please help to find what is causing this difference. Could any changes that went into iproute or kernel versions be causing this, or is any specific kernel configuration needs to be enabled for multipath with weight support? Both machines have CONFIG_IP_ROUTE_MULTIPATH=y.

Configuration:

    # ip link add link eth0 name eth0.100 type vlan id 100
    # ip link add link eth0 name eth0.200 type vlan id 200
    # ip link set eth0.100 up
    # ip link set eth0.200 up
    # ip -6 addr add 20::1/120 dev eth0.100
    # ip -6 addr add 30::1/120 dev eth0.200
    # ip -6 route add 40::/120 nexthop via 20::2 weight 10 nexthop via 30::2 weight 30

From 4.14:

    # ip -6 r l
    20::/120 dev eth0.100 proto kernel metric 256 pref medium
    30::/120 dev eth0.200 proto kernel metric 256 pref medium
    40::/120 metric 1024 pref medium
            nexthop via 20::2 dev eth0.100 weight 1
            nexthop via 30::2 dev eth0.200 weight 1
    
    # ip -V
    ip utility, iproute2-ss200330
    
    # uname -r
    4.14.231-pc64

From 4.19:

    # ip -6 r l
    ::1 dev lo proto kernel metric 256 pref medium
    20::/120 dev eth0.100 proto kernel metric 256 pref medium
    30::/120 dev eth0.200 proto kernel metric 256 pref medium
    40::/120 metric 1024
            nexthop via 20::2 dev eth0.100 weight 10
            nexthop via 30::2 dev eth0.200 weight 30 pref medium
    
    #  ip -V
    ip utility, iproute2-ss190107
    #  uname -r
    4.19.0-16-cloud-amd64
107
0 + 0
ธงชาติไทย
Elon Musk
I sit in a Tesla and translated this thread with Ai:

mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.