Incorrect netmask in network configuration

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • EN1GMA
    Confirmed User
    • Nov 2005
    • 1252

    #1

    Incorrect netmask in network configuration

    Hi guys,

    I had some problems with netmask of my additional IPs that was showing 255.255.255.0 instead of 255.255.255.255.
    I fixed that and contacted my hosting provider support to confirm if everything was correct with the new IP configuration showing the log of the command "ifconfig":

    root@server [~]# ifconfig
    eth0 Link encap:Ethernet HWaddr 00:22:15:AB:96:62
    inet addr:46.105.121.215 Bcast:46.105.121.255 Mask:255.255.255.0
    inet6 addr: fe80::222:15ff:feab:9662/64 Scope:Link
    UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
    RX packets:24879514 errors:0 dropped:0 overruns:0 frame:0
    TX packets:37628599 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:1000
    RX bytes:2963209878 (2.7 GiB) TX bytes:45812951640 (42.6 GiB)
    Interrupt:20 Memory:fe500000-fe520000

    eth0:cp1 Link encap:Ethernet HWaddr 00:22:15:AB:96:62
    inet addr:94.23.78.68 Bcast:94.23.78.255 Mask:255.255.255.255
    UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
    Interrupt:20 Memory:fe500000-fe520000

    eth0:0 Link encap:Ethernet HWaddr 00:22:15:AB:96:62
    inet addr:178.32.52.69 Bcast:178.32.52.69 Mask:255.255.255.255
    UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
    Interrupt:20 Memory:fe500000-fe520000

    lo Link encap:Local Loopback
    inet addr:127.0.0.1 Mask:255.0.0.0
    inet6 addr: ::1/128 Scope:Host
    UP LOOPBACK RUNNING MTU:16436 Metric:1
    RX packets:1495849 errors:0 dropped:0 overruns:0 frame:0
    TX packets:1495849 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:0
    RX bytes:697238017 (664.9 MiB) TX bytes:697238017 (664.9 MiB)

    And they said this:
    "Regarding the interface 'eth0:0', everything seems to be correct now. However, the interface 'eth0:cp1' IP broadcast is incorrect. As IP has a subnet mask '/32' the broadcast IP should be the IP itself (94.23.78.68)."
    I tried to fix by running the command:
    Code:
    sed -i 's/94.23.78.255/94.23.78.68/g' /etc/ips; /etc/init.d/ipaliases restart
    But I think that something is wrong.
    Here goes the log:
    root@server [~]# sed -i 's/94.23.78.255/94.23.78.68/g' /etc/ips; /etc/init.d/ipaliases restart
    Shutting down eth0:0 [ OK ]
    UnRouting 94.23.78.68 [ OK ]
    Shutting down eth0:cp1 [ OK ]
    Bringing up eth0:cp1 [ OK ]
    Routing 94.23.78.68 [ OK ]
    Bringing up eth0:cp2 [ OK ]
    Routing 178.32.52.69 [ OK ]

    root@server [~]# ifconfig
    eth0 Link encap:Ethernet HWaddr 00:22:15:AB:96:62
    inet addr:46.105.121.215 Bcast:46.105.121.255 Mask:255.255.255.0
    inet6 addr: fe80::222:15ff:feab:9662/64 Scope:Link
    UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
    RX packets:26237619 errors:0 dropped:0 overruns:0 frame:0
    TX packets:39680507 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:1000
    RX bytes:3122661542 (2.9 GiB) TX bytes:48244727511 (44.9 GiB)
    Interrupt:20 Memory:fe500000-fe520000

    eth0:cp1 Link encap:Ethernet HWaddr 00:22:15:AB:96:62
    inet addr:94.23.78.68 Bcast:94.23.78.68 Mask:255.255.255.255
    UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
    Interrupt:20 Memory:fe500000-fe520000

    eth0:cp2 Link encap:Ethernet HWaddr 00:22:15:AB:96:62
    inet addr:178.32.52.69 Bcast:178.32.52.255 Mask:255.255.255.255
    UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
    Interrupt:20 Memory:fe500000-fe520000

    lo Link encap:Local Loopback
    inet addr:127.0.0.1 Mask:255.0.0.0
    inet6 addr: ::1/128 Scope:Host
    UP LOOPBACK RUNNING MTU:16436 Metric:1
    RX packets:1571468 errors:0 dropped:0 overruns:0 frame:0
    TX packets:1571468 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:0
    RX bytes:725252851 (691.6 MiB) TX bytes:725252851 (691.6 MiB)
    I notice two things:

    1- I don't have 'eth0:0' anymore,now I have a 'eth0:cp1' and 'eth0:cp2', it seems that was renamed. Is this ok?

    2- It seems that eth0:cp1 is fine now:
    inet addr:94.23.78.68 Bcast:94.23.78.68 Mask:255.255.255.255

    But eth0:cp2 (old eth0:0 that was already fine) is not:
    inet addr:178.32.52.69 Bcast:178.32.52.255 Mask:255.255.255.255


    Any ideas? Can someone help me with these?

    Thank you guys!

    Regards!!
  • AMSM

    #2
    Why would you even need a netmask.

    Comment

    • Spudstr
      Confirmed User
      • Jan 2003
      • 2321

      #3
      Originally posted by EN1GMA
      Hi guys,

      I had some problems with netmask of my additional IPs that was showing 255.255.255.0 instead of 255.255.255.255.
      I fixed that and contacted my hosting provider support to confirm if everything was correct with the new IP configuration showing the log of the command "ifconfig":




      And they said this:


      I tried to fix by running the command:
      Code:
      sed -i 's/94.23.78.255/94.23.78.68/g' /etc/ips; /etc/init.d/ipaliases restart
      But I think that something is wrong.
      Here goes the log:


      I notice two things:

      1- I don't have 'eth0:0' anymore,now I have a 'eth0:cp1' and 'eth0:cp2', it seems that was renamed. Is this ok?

      2- It seems that eth0:cp1 is fine now:
      inet addr:94.23.78.68 Bcast:94.23.78.68 Mask:255.255.255.255

      But eth0:cp2 (old eth0:0 that was already fine) is not:
      inet addr:178.32.52.69 Bcast:178.32.52.255 Mask:255.255.255.255


      Any ideas? Can someone help me with these?

      Thank you guys!

      Regards!!
      According to documentation..
      These are subinterfaces and a method of adding additional IP addresses along with their subnets to an interface. The part after the colon is arbitrary.

      iptables cannot reference subinterfaces, so any rules required would be based on IP addresses, but bound to the physical interface if required.

      As far as netmasks go. having a /32 netmask on an alias ip is fine actually. since the only one that matters is the netmask that defines the default route which is accessible is the main IP on the box, well you could use any actually as long as your default gateway is pointing to an aces sable IP on that netmask. Since they are different ips not in the same broadcast domain issuing a /32 on each IP is the correct way to do this masking.

      Now with that being said this is actually HORRIFIC practice and VERY poor network design which will lead into a lot of "shit talking" on the broadcast domain, lots of crap traffic and actually you are vulnerable to a broadcast flood on the broadcast domain, which would take you down and everyone else on that broadcast domain if it was to happen.

      In short this his horrible, while I realize this is OVH.. its not a real good design to do.
      Managed Hosting - Colocation - Network Services
      Yellow Fiber Networks
      icq: 19876563

      Comment

      • EN1GMA
        Confirmed User
        • Nov 2005
        • 1252

        #4
        Spudstr, thank you for your reply buddy.

        So what do you suggest to do?

        Thanks once again!

        Comment

        • Spudstr
          Confirmed User
          • Jan 2003
          • 2321

          #5
          Originally posted by EN1GMA
          Spudstr, thank you for your reply buddy.

          So what do you suggest to do?

          Thanks once again!
          Stick with the /32's and don't change the netmasks on them.

          Unless they just don't work? then that is a different set of issues.
          Managed Hosting - Colocation - Network Services
          Yellow Fiber Networks
          icq: 19876563

          Comment

          Working...