2NICs configured, but used only one

  • Hello everybody.


    I faced strange behavior of newly installed and fully updated OMV on host with 2 NICs - integrated on S5000PSL board.
    Both NICs configured static in one subnet, for example:
    eth0 - 192.168.1.1
    eth1 - 192.168.1.2
    no gateway configured.


    From my PC (192.168.1.3) I start pinging eth0 and eth1.


    Now I take off eth0 cable - ping stopped to both(!) interfaces. Connect cable back to eth0 - both ping started response.
    I take off eth1 cable - ping go on on both(!) interfaces.


    Arping to eth0 and eth1 return MAC-address of eth0 in both cases.


    Am I missed something in configuration or there is a bug in OMV or Debian?


    Same behavior mentioned here http://forums.openmediavault.org/viewtopic.php?f=15&t=4031

  • Did you try to restart the ping after pulling the cable? I think the ping mechanism stores the mac of the target device...


    Greetings
    David

    "Well... lately this forum has become support for everything except omv" [...] "And is like someone is banning Google from their browsers"


    Only two things are infinite, the universe and human stupidity, and I'm not sure about the former.

    Upload Logfile via WebGUI/CLI
    #openmediavault on freenode IRC | German & English | GMT+1
    Absolutely no Support via PM!

  • No, I didn't, but described situation is result of tests, after i found such behavior on:
    OMV configuration from first post (iSCSI target) + XenServer 6.2 (iSCSI initiator). Just after configuration XS shows 2 active path to OMV.
    Disconnect cable from eth0 - comletely no connection to OMV - XS, browser from my PC to OMV WebUI, ping from my PC.
    Reconnect cable - all connections restored
    I'll test restart ping, but not sure it helps.

  • Just installed OMV 0.5.0.24 (without any updates, clean install) to VirtualBox with 2 NICs, connected to internal network. eth0 - 172.16.1.10, eth1 - 172.16.1.20.
    Management PC - Lubuntu at same internal network on Virtualbox, eth0 - 172.16.1.50.
    Run 2 terminals with ping to *.10 and *.20 - same behavior.


    Its look like eth0 is primary interface and if it unavailable - no any connection to OMV box. May be its routing issue.


    About arping:
    First responce from each NIC with MAC of eth1, second responce and all other - with MAC of eth0. No matter which NIC I arping.


    Connect/disconnect cable from virtual NICs has same behavior with arping, as with ping.

  • It seems, that no idea what's wrong with 2 NIC configuration...


    Some more test info: if configured two different subnets on interfaces - all work as expected. Disconnecting cable from any NIC do not affect other NIC connection.
    But i need two NICs on other side


    For me acceptable solution is to configure bond interface. it also works as expected and do not break connection to host, while one cable is disconected from any NIC. I tried LACP configuration on host and on switch.

Jetzt mitmachen!

Sie haben noch kein Benutzerkonto auf unserer Seite? Registrieren Sie sich kostenlos und nehmen Sie an unserer Community teil!