Netgate Discussion Forum
    • Categories
    • Recent
    • Tags
    • Popular
    • Users
    • Search
    • Register
    • Login

    HA CARP - IPv6 Two masters

    Scheduled Pinned Locked Moved HA/CARP/VIPs
    56 Posts 11 Posters 14.9k Views 6 Watching
    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.
    • DerelictD Offline
      Derelict LAYER 8 Netgate
      last edited by

      Nice catch.

      LAN@213  fd57:187e:523f:0715::f    MASTER

      Chattanooga, Tennessee, USA
      A comprehensive network diagram is worth 10,000 words and 15 conference calls.
      DO NOT set a source address/port in a port forward or firewall rule unless you KNOW you need it!
      Do Not Chat For Help! NO_WAN_EGRESS(TM)

      1 Reply Last reply Reply Quote 0
      • I Offline
        IcePick
        last edited by

        I am having exactly the same as this since moving to 2.4 from 2.3.5
        interesting only on 2 of the 4 IPv6 CARPs
        they were the only 2 that could use :: in their address
        I tried expanding to 0:0:0:
        it did not help

        I have confirmed by packet capture that packets to ff02::12 are seen on both systems

        –------------
        Ok I figured out how to get it to a normal state (all master on primary and all backup on secondary).
        You need to reboot the backup firewall, and while its rebooting clear the firewall states on the primary.
        Carp failover works perfectly when its like this but there is still an issue.

        ANY configuration sync (manual/auto) from the primary to the backup causes the backup to become master on the two IPV6 carps.

        1 Reply Last reply Reply Quote 0
        • I Offline
          IcePick
          last edited by

          To reiterate I did not have this issue until upgrading from 2.3.5 to 2.4.2-RELEASE-p1, or at least it seemed to have gotten worse.

          More testing:
          changed from x::1 (ie X:0:0:0:1)  to x:1:1:1:1
          on one of the CARP interfaces and the problem went away
          Did not change the real interface IP

          UPDATE: it worked for the first one, but broke both after I changed the second one.
          Why are these 2 different then the other 2?
          they connect to the same switch
          I found a difference, one set of addresses used all lower case for the hex in the address, the none working ones had capitals.
          I have changed all to lower and rebooted B unit and it came up all in backup, did not have to reset states on A firewall.
          I'm not saying this is the issue - but giving people ideas of what I found
          So in summary: using all lower case for hex and changed the addresses to ones that can not condense to ::

          1 Reply Last reply Reply Quote 0
          • A Offline
            anthonysomerset
            last edited by

            Just want to add i appear to have hit this "bug" in one of our SG-4860 clusters

            our IPv6 addresses are in their shortened form with no leading zeros, had to reboot secondary to clear this out, will keep an eye on things

            1 Reply Last reply Reply Quote 0
            • X Offline
              xciter327
              last edited by

              I am also hitting something similar this in our office/test system.

              Both devices are connected to a Cisco 3560G switch. IGMP snooping and ipv6 mld snooping are disabled. All ports are set to "portfast". There are no "loops" in the network. There are no topology changes.

              You will notice that each one sees the others advertisements and their own.

              Primary:
              16:42:40.428976 IP6 fe80::ec4:7aff:feab:3724 > ff02::12: ip-proto-112 36
              16:42:42.597228 IP6 fe80::ec4:7aff:feac:821a > ff02::12: ip-proto-112 36
              16:42:50.886692 IP6 fe80::ec4:7aff:feab:3724 > ff02::12: ip-proto-112 36
              16:42:52.607533 IP6 fe80::ec4:7aff:feac:821a > ff02::12: ip-proto-112 36
              16:43:01.382988 IP6 fe80::ec4:7aff:feab:3724 > ff02::12: ip-proto-112 36
              16:43:02.612549 IP6 fe80::ec4:7aff:feac:821a > ff02::12: ip-proto-112 36

              Backup:
              16:42:09.212760 IP6 fe80::ec4:7aff:feab:3724 > ff02::12: ip-proto-112 36
              16:42:12.573960 IP6 fe80::ec4:7aff:feac:821a > ff02::12: ip-proto-112 36
              16:42:19.608720 IP6 fe80::ec4:7aff:feab:3724 > ff02::12: ip-proto-112 36
              16:42:22.578900 IP6 fe80::ec4:7aff:feac:821a > ff02::12: ip-proto-112 36
              16:42:30.015028 IP6 fe80::ec4:7aff:feab:3724 > ff02::12: ip-proto-112 36
              16:42:32.585911 IP6 fe80::ec4:7aff:feac:821a > ff02::12: ip-proto-112 36

              This only happens for IPv6 CARP IPs.

              Here are the interfaces, just to confirm the vhid:

              Primary:
              igb0: flags=8943 <up,broadcast,running,promisc,simplex,multicast>metric 0 mtu 1500
              options=6400bb <rxcsum,txcsum,vlan_mtu,vlan_hwtagging,jumbo_mtu,vlan_hwcsum,vlan_hwtso,rxcsum_ipv6,txcsum_ipv6>ether 0c:c4:7a:ac:82:1a
              hwaddr 0c:c4:7a:ac:82:1a
              inet6 fe80::ec4:7aff:feac:821a%igb0 prefixlen 64 scopeid 0x1
              inet6 xxxx:xxxx:1:2::3 prefixlen 124
              inet6 xxxx:xxxx:1:2::2 prefixlen 124 vhid 4
              inet yyy.yyy.233.108 netmask 0xfffffff0 broadcast yyy.yyy.233.111
              inet yyy.yyy.233.110 netmask 0xfffffff0 broadcast yyy.yyy.233.111 vhid 1
              nd6 options=21 <performnud,auto_linklocal>media: Ethernet autoselect (1000baseT <full-duplex>)
              status: active
              carp: MASTER vhid 1 advbase 10 advskew 1
              carp: MASTER vhid 4 advbase 10 advskew 1

              Backup:
              igb0: flags=8943 <up,broadcast,running,promisc,simplex,multicast>metric 0 mtu 1500
              options=6400bb <rxcsum,txcsum,vlan_mtu,vlan_hwtagging,jumbo_mtu,vlan_hwcsum,vlan_hwtso,rxcsum_ipv6,txcsum_ipv6>ether 0c:c4:7a🆎37:24
              hwaddr 0c:c4:7a🆎37:24
              inet6 fe80::ec4:7aff:feab:3724%igb0 prefixlen 64 scopeid 0x1
              inet6 xxxx:xxxx:1:2::4 prefixlen 124
              inet yyy.yyy.233.109 netmask 0xfffffff0 broadcast yyy.yyy.233.111
              inet yyy.yyy.233.110 netmask 0xfffffff0 broadcast yyy.yyy.233.111 vhid 1
              nd6 options=21 <performnud,auto_linklocal>media: Ethernet autoselect (1000baseT <full-duplex>)
              status: active
              carp: MASTER vhid 4 advbase 10 advskew 101
              carp: BACKUP vhid 1 advbase 10 advskew 101

              So the CARP interface is correctly assigned to the primary node, but the backup one still claims its master in the dashboard and with "ifconfig igb0".</full-duplex></performnud,auto_linklocal></rxcsum,txcsum,vlan_mtu,vlan_hwtagging,jumbo_mtu,vlan_hwcsum,vlan_hwtso,rxcsum_ipv6,txcsum_ipv6></up,broadcast,running,promisc,simplex,multicast></full-duplex></performnud,auto_linklocal></rxcsum,txcsum,vlan_mtu,vlan_hwtagging,jumbo_mtu,vlan_hwcsum,vlan_hwtso,rxcsum_ipv6,txcsum_ipv6></up,broadcast,running,promisc,simplex,multicast>

              1 Reply Last reply Reply Quote 0
              • DerelictD Offline
                Derelict LAYER 8 Netgate
                last edited by Derelict

                Why did you play with advbase/advskew?

                Use 1/0 on the primary that will sync 1/100 to the secondary. Then just leave it alone.

                Chattanooga, Tennessee, USA
                A comprehensive network diagram is worth 10,000 words and 15 conference calls.
                DO NOT set a source address/port in a port forward or firewall rule unless you KNOW you need it!
                Do Not Chat For Help! NO_WAN_EGRESS(TM)

                1 Reply Last reply Reply Quote 0
                • X Offline
                  xciter327
                  last edited by

                  Yes. I did try multiple base values between 0 - 20 for the base and 0 and 1 for skew. The settings are correctly(+100 for skew) transferred to the backup unit. Still backup thinks it's primary for IPv6.

                  1 Reply Last reply Reply Quote 0
                  • DerelictD Offline
                    Derelict LAYER 8 Netgate
                    last edited by Derelict

                    Are you 100% certain the case described in reply #15 ^ is not present?

                    Use 1/0 on the primary that will sync 1/100 to the secondary. Then just leave it alone.

                    Just do that. If changing it didn't correct it it is not the problem.

                    Packet capture on both nodes and see if you see the CARP going out the interface or in the interface. You can filter on CARP only in Diagnostics > Packet Capture.

                    Chattanooga, Tennessee, USA
                    A comprehensive network diagram is worth 10,000 words and 15 conference calls.
                    DO NOT set a source address/port in a port forward or firewall rule unless you KNOW you need it!
                    Do Not Chat For Help! NO_WAN_EGRESS(TM)

                    1 Reply Last reply Reply Quote 0
                    • X Offline
                      xciter327
                      last edited by

                      1. Regarding post #15 solution. I tried both shorthand(no leading zeroes) and full notation with nothing omitted.
                      2. I included a tcpdump in my first post. It looks to me that they are both receiving each other's updates.

                      1 Reply Last reply Reply Quote 0
                      • I Offline
                        IcePick
                        last edited by

                        Have you tried changing to addresses that CAN NOT be shortened to have a :: ?

                        1 Reply Last reply Reply Quote 0
                        • X Offline
                          xciter327
                          last edited by

                          @IcePick:

                          Have you tried changing to addresses that CAN NOT be shortened to have a :: ?

                          Yes I did. No difference.

                          1 Reply Last reply Reply Quote 0
                          • DerelictD Offline
                            Derelict LAYER 8 Netgate
                            last edited by

                            Did you put base/skew back to the default or not?

                            Chattanooga, Tennessee, USA
                            A comprehensive network diagram is worth 10,000 words and 15 conference calls.
                            DO NOT set a source address/port in a port forward or firewall rule unless you KNOW you need it!
                            Do Not Chat For Help! NO_WAN_EGRESS(TM)

                            1 Reply Last reply Reply Quote 0
                            • X Offline
                              xciter327
                              last edited by

                              @Derelict:

                              Did you put base/skew back to the default or not?

                              Yes, I did.

                              1 Reply Last reply Reply Quote 0
                              • DerelictD Offline
                                Derelict LAYER 8 Netgate
                                last edited by

                                Well, cut loose with more. Screen shots, pcaps, whatever. IPv6 CARP works.

                                Chattanooga, Tennessee, USA
                                A comprehensive network diagram is worth 10,000 words and 15 conference calls.
                                DO NOT set a source address/port in a port forward or firewall rule unless you KNOW you need it!
                                Do Not Chat For Help! NO_WAN_EGRESS(TM)

                                1 Reply Last reply Reply Quote 0
                                • X Offline
                                  xciter327
                                  last edited by

                                  I disabled "DHCP Snooping" on the directly connected switch. That was somehow blocking stuff. Seems to be working OK now. I can no longer reproduce the issue. Will post if I can.

                                  1 Reply Last reply Reply Quote 0
                                  • DerelictD Offline
                                    Derelict LAYER 8 Netgate
                                    last edited by

                                    Amazing. It was a setting on the switch. Simply amazing.

                                    Glad you found it.

                                    Chattanooga, Tennessee, USA
                                    A comprehensive network diagram is worth 10,000 words and 15 conference calls.
                                    DO NOT set a source address/port in a port forward or firewall rule unless you KNOW you need it!
                                    Do Not Chat For Help! NO_WAN_EGRESS(TM)

                                    1 Reply Last reply Reply Quote 0
                                    • X Offline
                                      xciter327
                                      last edited by

                                      So the problem is kind of back.

                                      Same situation. Secondary pfsense become master for both IPv6 CARP groups, both report as master. The weird thing now is that if I shut down the secondary pfsense box IPv6 stops working completely. The primary box reports CARP status "Master"(as it always does), but the address is not reachable on the local LAN.

                                      IGMP / DHCP snooping is disabled on the two switches between test PC and firewalls. The IPv4 CARP works fine.

                                      1 Reply Last reply Reply Quote 0
                                      • DerelictD Offline
                                        Derelict LAYER 8 Netgate
                                        last edited by

                                        Again, it sounds like something at layer 2.

                                        Either of the nodes will show MASTER if it does not receive the heartbeats from the other node. Solving dual MASTER is generally as simple as fixing the reason(s) that one node is not seeing the heartbeats from the other node.

                                        Chattanooga, Tennessee, USA
                                        A comprehensive network diagram is worth 10,000 words and 15 conference calls.
                                        DO NOT set a source address/port in a port forward or firewall rule unless you KNOW you need it!
                                        Do Not Chat For Help! NO_WAN_EGRESS(TM)

                                        1 Reply Last reply Reply Quote 0
                                        • X Offline
                                          xciter327
                                          last edited by xciter327

                                          It most definitely not L2 issue. The devices can see each other. Confirmed with tcpdump.( tcpdump -i igb0 -ttt -n proto CARP). There are 2 VHIDs on this interface.

                                          Master:
                                          00:00:00.000094 IP 217.117.yyy.xxx > 224.0.0.18: VRRPv2, Advertisement, vrid 1, prio 1, authtype none, intvl 1s, length 36
                                          00:00:01.004961 IP 217.117.yyy.xxx > 224.0.0.18: VRRPv2, Advertisement, vrid 1, prio 1, authtype none, intvl 1s, length 36
                                          00:00:00.000103 IP6 fe80::ec4:7aff:feac:821a > ff02::12: ip-proto-112 36
                                          00:00:01.005069 IP6 fe80::ec4:7aff:feac:821a > ff02::12: ip-proto-112 36

                                          Backup:
                                          00:00:00.000064 IP 217.117.yyy.xxx > 224.0.0.18: VRRPv2, Advertisement, vrid 1, prio 1, authtype none, intvl 1s, length 36
                                          00:00:01.004781 IP 217.117.yyy.xxx > 224.0.0.18: VRRPv2, Advertisement, vrid 1, prio 1, authtype none, intvl 1s, length 36
                                          00:00:00.000062 IP6 fe80::ec4:7aff:feac:821a > ff02::12: ip-proto-112 36
                                          00:00:01.004907 IP6 fe80::ec4:7aff:feac:821a > ff02::12: ip-proto-112 36

                                          What I did is to download a config backup from each unit and a do a restore from config. That fixed the issue for me. There were no changes made to the underlying switching network.

                                          I also hit this bug with 2.4.3-Release-P1. That left me with one extra VHID on each interface stuck in "INIT" state. Rebooting the firewall is the only way I found to fix it.

                                          P.S. - Yesterday I also tried shutting down the "Backup" unit and fully un-plugging it from the network. While the IPv6 CARP interface on the LAN was showing as "up" and "master" on the only firewall left, IPv6 connectivity was not working until I rebooted the firewall.

                                          1 Reply Last reply Reply Quote 0
                                          • X Offline
                                            xciter327
                                            last edited by

                                            This post is deleted!
                                            1 Reply Last reply Reply Quote 0
                                            • First post
                                              Last post
                                            Copyright 2025 Rubicon Communications LLC (Netgate). All rights reserved.