ML
    • Recent
    • Categories
    • Tags
    • Popular
    • Users
    • Groups
    • Register
    • Login

    CentOS 7 networking issues

    IT Discussion
    centos centos 7 networking
    4
    16
    3.2k
    Loading More Posts
    • Oldest to Newest
    • Newest to Oldest
    • Most Votes
    Reply
    • Reply as topic
    Log in to reply
    This topic has been deleted. Only users with topic management privileges can see it.
    • JaredBuschJ
      JaredBusch
      last edited by

      So I go and delete eth0 from nmtui and then systemctl restart network

      Everything still works, but the old file is gone.

      [root@nginix ~]# ls /etc/sysconfig/network-scripts/ifcfg-eth0
      ls: cannot access /etc/sysconfig/network-scripts/ifcfg-eth0: No such file or directory
      
      1 Reply Last reply Reply Quote 0
      • JaredBuschJ
        JaredBusch
        last edited by

        I then edited "Wired connection 1" to show this.
        0_1488408575453_upload-f6b7cd4f-26c1-4408-bf77-ffdeaeb06a78
        0_1488408598656_upload-20e5aea3-4ee3-4798-b693-8bf5cbaf342e

        It all works.

        [root@nginix ~]# nmtui
        [root@nginix ~]# systemctl restart network
        [root@nginix ~]# ip a sh
        1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1
            link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
            inet 127.0.0.1/8 scope host lo
               valid_lft forever preferred_lft forever
            inet6 ::1/128 scope host
               valid_lft forever preferred_lft forever
        2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
            link/ether 02:15:5d:01:03:09 brd ff:ff:ff:ff:ff:ff
            inet 10.201.1.18/24 brd 10.201.1.255 scope global dynamic eth0
               valid_lft 86396sec preferred_lft 86396sec
            inet6 fe80::fb6c:e1ec:b116:5220/64 scope link
               valid_lft forever preferred_lft forever
        

        But now there is a legacy file again.

        [root@nginix ~]# ls /etc/sysconfig/network-scripts/ifcfg-eth0
        /etc/sysconfig/network-scripts/ifcfg-eth0
        [root@nginix ~]# cat /etc/sysconfig/network-scripts/ifcfg-eth0
        HWADDR=02:15:5D:01:03:09
        TYPE=Ethernet
        BOOTPROTO=dhcp
        DEFROUTE=yes
        PEERDNS=yes
        PEERROUTES=yes
        IPV4_FAILURE_FATAL=no
        IPV6INIT=yes
        IPV6_AUTOCONF=yes
        IPV6_DEFROUTE=yes
        IPV6_PEERDNS=yes
        IPV6_PEERROUTES=yes
        IPV6_FAILURE_FATAL=no
        IPV6_ADDR_GEN_MODE=stable-privacy
        NAME=eth0
        UUID=a7cf382b-8b9f-3abe-a3fe-2d4e8490a666
        ONBOOT=yes
        AUTOCONNECT_PRIORITY=-999
        
        1 Reply Last reply Reply Quote 0
        • JaredBuschJ
          JaredBusch
          last edited by JaredBusch

          and nmcli device show is different.

          [root@nginix ~]# nmcli device show
          GENERAL.DEVICE:                         eth0
          GENERAL.TYPE:                           ethernet
          GENERAL.HWADDR:                         02:15:5D:01:03:09
          GENERAL.MTU:                            1500
          GENERAL.STATE:                          100 (connected)
          GENERAL.CONNECTION:                     eth0
          GENERAL.CON-PATH:                       /org/freedesktop/NetworkManager/ActiveConnection/1
          WIRED-PROPERTIES.CARRIER:               on
          IP4.ADDRESS[1]:                         10.201.1.18/24
          IP4.GATEWAY:                            10.201.1.1
          IP4.DNS[1]:                             10.201.1.7
          IP4.DOMAIN[1]:                          domain.local
          IP6.ADDRESS[1]:                         fe80::fb6c:e1ec:b116:5220/64
          IP6.GATEWAY:
          
          GENERAL.DEVICE:                         lo
          GENERAL.TYPE:                           loopback
          GENERAL.HWADDR:                         00:00:00:00:00:00
          GENERAL.MTU:                            65536
          GENERAL.STATE:                          10 (unmanaged)
          GENERAL.CONNECTION:                     --
          GENERAL.CON-PATH:                       --
          IP4.ADDRESS[1]:                         127.0.0.1/8
          IP4.GATEWAY:
          IP6.ADDRESS[1]:                         ::1/128
          IP6.GATEWAY:
          
          1 Reply Last reply Reply Quote 0
          • ObsolesceO
            Obsolesce
            last edited by

            First thing I do on CentOS is disable Network Manager. I hate it.

            JaredBuschJ 1 Reply Last reply Reply Quote 0
            • JaredBuschJ
              JaredBusch @Obsolesce
              last edited by

              @Tim_G said in CentOS 7 networking issues:

              First thing I do on CentOS is disable Network Manager. I hate it.

              Wrong answer, sorry. Network manager is the new default. You should get used to it and use it appropriately.

              1 Reply Last reply Reply Quote 2
              • JaredBuschJ
                JaredBusch
                last edited by

                So I changed it to static like this.

                0_1488409027886_upload-6278cf40-5144-4ad9-a5ed-a3593f25ef3e

                1 Reply Last reply Reply Quote 0
                • ObsolesceO
                  Obsolesce
                  last edited by

                  Part of my CentOS routine:

                  1. Change run level: nano /etc/inittab (use id:3:initdefault:)

                  2. cd /etc/rc.d/rc3.d

                  3. ls -lh

                  4. mv S23NetworkManager K23NetworkManager

                  5. Reboot

                  6. ls -l /etc/sysconfig/network-scripts (to see what's there)

                  7. nano /etc/sysconfig/network-scripts/ifcfg-eth0
                    (if dhcp)
                    DEVICE=eth0
                    BOOTPROTO=dhcp
                    ONBOOT=yes
                    (save it)

                  ifup eth0
                  chkconfig network on

                  stacksofplatesS 1 Reply Last reply Reply Quote 0
                  • travisdh1T
                    travisdh1
                    last edited by

                    Network Manager, still throwing admins for a loop even after all these years 😞

                    1 Reply Last reply Reply Quote 0
                    • stacksofplatesS
                      stacksofplates
                      last edited by stacksofplates

                      NetworkManager still creates the ifcfg files under network-scripts. So those configs will still be there, but should have NM_CONTROLLED=yes

                      Also, for some reason CentOS ships with both network and NetworkManager on at the same time. So that's most likely why you saw the DHCP for NetworkManager and the static file that you created. I always disable and mask network to keep people from turning it back on when they aren't paying attention.

                      stacksofplatesS 1 Reply Last reply Reply Quote 0
                      • stacksofplatesS
                        stacksofplates @Obsolesce
                        last edited by stacksofplates

                        @Tim_G said in CentOS 7 networking issues:

                        Part of my CentOS routine:

                        1. Change run level: nano /etc/inittab (use id:3:initdefault:)

                        2. cd /etc/rc.d/rc3.d

                        3. ls -lh

                        4. mv S23NetworkManager K23NetworkManager

                        5. Reboot

                        6. ls -l /etc/sysconfig/network-scripts (to see what's there)

                        7. nano /etc/sysconfig/network-scripts/ifcfg-eth0
                          (if dhcp)
                          DEVICE=eth0
                          BOOTPROTO=dhcp
                          ONBOOT=yes
                          (save it)

                        ifup eth0
                        chkconfig network on

                        Targets (run levels) are handled through systemd now so you should use systemctl set-default multi-user.target.

                        And as @JaredBusch said, NetworkManager is the new default, and handles things like teaming much better.

                        ObsolesceO 1 Reply Last reply Reply Quote 1
                        • ObsolesceO
                          Obsolesce @stacksofplates
                          last edited by

                          @stacksofplates Didn't realize he was being forced to use defaults. Since that's the case, I guess he has no other option than to use Network Manager?

                          JaredBuschJ 1 Reply Last reply Reply Quote 0
                          • JaredBuschJ
                            JaredBusch @Obsolesce
                            last edited by

                            @Tim_G said in CentOS 7 networking issues:

                            @stacksofplates Didn't realize he was being forced to use defaults. Since that's the case, I guess he has no other option than to use Network Manager?

                            No one is forcing me to use anything. But use of defaults is a best practice. without a really good reason, you do not stray from them.

                            ObsolesceO 1 Reply Last reply Reply Quote 2
                            • ObsolesceO
                              Obsolesce @JaredBusch
                              last edited by

                              @JaredBusch said in CentOS 7 networking issues:

                              @Tim_G said in CentOS 7 networking issues:

                              @stacksofplates Didn't realize he was being forced to use defaults. Since that's the case, I guess he has no other option than to use Network Manager?

                              No one is forcing me to use anything. But use of defaults is a best practice. without a really good reason, you do not stray from them.

                              Losing VM functionality due to time issues because of a default isn't a good enough reason? Isn't it a production VM? Does it being down not lose the company money on top of paying you to mess with it? I'd rather use a non-default that works just as well for a particular function or task, rather than having to waste so much of a clients money to fix something that can arguably be replaced with something else non-default. Maybe Network Manager handles teaming better, but are you using teaming? Are you using teaming at the VM level or at the Hypervisor level? Just do what's best for the client. Don't use a default because it does something better with something you aren't even using. Know what I mean?

                              Maybe for your case it's worth it to get Network Manager working properly due to other things I don't know about. I'm not giving a blanket statement, just throwing out a point is all.

                              Sometimes best practice is a square, and what you are dealing with is a circle. It might not fit.

                              stacksofplatesS 1 Reply Last reply Reply Quote 0
                              • stacksofplatesS
                                stacksofplates @Obsolesce
                                last edited by

                                @Tim_G said in CentOS 7 networking issues:

                                Losing VM functionality due to time issues because of a default isn't a good enough reason

                                That's not the reason. The downtime was due to both services running. I have literally never had a network issue with just NetworkManager running. It also takes about 10 seconds to set up an interface through NetworkManager, it's much simpler.

                                Also network is being phased out, and not receiving feature updates. So while you can use it currently, it's most likely not going to be there in 8.

                                And yes, I am using teaming, but on the host.

                                1 Reply Last reply Reply Quote 0
                                • stacksofplatesS
                                  stacksofplates @stacksofplates
                                  last edited by

                                  @stacksofplates said in CentOS 7 networking issues:

                                  NetworkManager still creates the ifcfg files under network-scripts. So those configs will still be there, but should have NM_CONTROLLED=yes

                                  Also, for some reason CentOS ships with both network and NetworkManager on at the same time. So that's most likely why you saw the DHCP for NetworkManager and the static file that you created. I always disable and mask network to keep people from turning it back on when they aren't paying attention.

                                  So it looks like they took out the NM_CONTROLLED section. I must have been thinking of RHEL 6. But NM still creates the ifcfg files in the same directory.

                                  1 Reply Last reply Reply Quote 0
                                  • 1 / 1
                                  • First post
                                    Last post