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

    2.3 - LAGG, VLAN, Carp - after Update no route

    Scheduled Pinned Locked Moved General pfSense Questions
    6 Posts 3 Posters 1.3k Views
    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.
    • M Offline
      masterd01
      last edited by

      Hi,

      we have many PFSense-Installations and updated a lot of them up to 2.3. On two
      Servers we got really big problems. An importent on :)
      The network there:

      2x IBM Server x3650 (6x Network-IF, Two OnBoard, 4x on network cards) –> LAGG
      (Failover) --> VLAN --> CARP --> HASync and Config --> HaProxy/Loadbalancing

      The Problem:
      After Update the PFSense from 2.2.6 to 2.3 everything was fine. After Reboot the
      Master-Server the Failover goes to the Second. Everything is still fine. If the
      Master comes back, the Carp switches again but no traffic is routed between the
      Networks behind the PF. Ping to both PFs + Carp from all IFs are okay. From PF i
      could ping everything. But not over the PF.
      If i boot the Second one - there i got the Problem also. No network-connection over
      the PFSense. Sometimes it helps to open some Gateway-Settings (equal one) and safe
      it again.
      But this is not often a solution.

      What i've tested:

      • Reboot
      • Disable Carp
      • Default-Config PFSense (with Backup from mine)
      • PFSense 2.3.1
      • HAProxy reinstalled (with bugs, look attached)
      • HAProxy Dev
      • Add VLANs directly on one Network IF (disable LAGG, attached)
      • Delete LAGG and make it new (with a failure - look attached)
      • Delete all network-config and make it new (also attached)

      Nothing helps.

      On the 2.2.6 this config is okay. No Problems are known.

      Another Problem at these both Servers:
      The GUI is really slow. Sometimes i could only do one change, click safe and the GUI
      wait. And wait. Than i got Gateway-Timeout (look attached). I could resolv it with
      restart PHP-FPM (16) and Restarting WebConfigurator (11).

      Bevor i do a Rollback to 2.2.6. i ask you for help. Have i missed something?

      Thats for your time and help!

      MasterD

      lagg1.PNG
      lagg1.PNG_thumb
      timeout1.PNG
      timeout1.PNG_thumb
      if.PNG
      if.PNG_thumb
      vlans.PNG
      vlans.PNG_thumb
      lagg2.PNG
      lagg2.PNG_thumb
      ha-error.txt

      1 Reply Last reply Reply Quote 0
      • M Offline
        mevans336
        last edited by

        Have you manually changed the MTU on your lagg interfaces? If so, the settings under the INTERFACE may have been lost, even though the MTU is reflected properly under the Status - Interface page.

        1 Reply Last reply Reply Quote 0
        • M Offline
          masterd01
          last edited by

          Hi Mevans,

          no it's till at 1500.

          1 Reply Last reply Reply Quote 0
          • P Offline
            PiBa
            last edited by

            Remove the 'old' haproxy files they likely are messing up some stuff for the pf / nat rules as well.

            https://forum.pfsense.org/index.php?topic=109783.msg611641#msg611641

            Then reinstall the haproxy package, that should at least get rid of the haproxy errors, and its possible interference with rule loading. Not sure if that will solve the lagg interface issues though..

            1 Reply Last reply Reply Quote 0
            • M Offline
              masterd01
              last edited by

              Hi PiBa,

              that's for that hint.
              The HaProxy-Error-Message is gone after deleting everything.
              The LAGG-Problem is still there.

              I had to downgrade to 2.2.6. There issn't the Problem anymore.

              Any other ideas?

              1 Reply Last reply Reply Quote 0
              • M Offline
                masterd01
                last edited by

                Hi again,

                my Problem is still there. I found out now, that the Problem is the Slave-System!
                Exactly after five days the second Server does something with the Carp and the Routing failes.
                I don't know what happen there but after reboot from the Slave-System everything is fine again - till the next five days.

                The Master-Hardware is changed, the slave not. Should i?
                What should i Test next?
                I have no ideas anymore and it's not so nice to get sunday a wake up call from the company that the problem is back again.

                Thanks!

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