XCP-ng
    • Categories
    • Recent
    • Tags
    • Popular
    • Users
    • Groups
    • Register
    • Login

    Issue installing latest pfSense Plus (24.03 release)

    Scheduled Pinned Locked Moved Compute
    19 Posts 5 Posters 1.1k Views 4 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.
    • nikadeN Offline
      nikade Top contributor
      last edited by

      What happends if you create a new VM with the template "other" and attempt to make a clean installation? Do you get the same error?

      Also, what kind of configuration have to given the machine? For example NIC? RAM? Dynamic RAM or static?

      1 Reply Last reply Reply Quote 0
      • A Offline
        Affonso @AtaxyaNetwork
        last edited by

        @AtaxyaNetwork I'm using XCP-ng 8.2.1
        There isn't a 24.03 ISO as it is the paid (Plus) option.

        Right now to install I have to install the 2.7.2 CE version, then Netgate systems identify the Netgate ID of the device and allow the upgrade to pfSense 23.09.1 Plus version, and then you are offered the 24.03 Plus version. There isn't a direct update from 2.7.2 CE to 24.03 Plus.

        I have installed the 24.03 version on physical devices. However the virtualised devices, after the update and upon first boot, fail to boot as they crash.

        I'm not guy to diagnose boot crashes. I made the possible captures.
        All I can affirm without doubt is that the error is exactly the same on two completely different machines.

        @nikade they have always been created using the "other" template.
        The original VM's had 1 vCPU and 2 GB preset on the template. Aside from the values changed initially, I didn't made changes to the memory/resources.
        NIC: 2 NIC's (one for WAN another for LAN). Initially they always have the "Realtek" driver, so all I did on one test was changing that to Intel e1000. But there was no change to the outcome.

        nikadeN 1 Reply Last reply Reply Quote 0
        • nikadeN Offline
          nikade Top contributor @Affonso
          last edited by

          @nikade they have always been created using the "other" template.
          The original VM's had 1 vCPU and 2 GB preset on the template. Aside from the values changed initially, I didn't made changes to the memory/resources.
          NIC: 2 NIC's (one for WAN another for LAN). Initially they always have the "Realtek" driver, so all I did on one test was changing that to Intel e1000. But there was no change to the outcome.

          Hi,

          Can you go to the "Advanced" tab of the VM and show the memory settings?
          I want to make sure you're not using dynamic memory since it is really unstable in Xen.

          8cb46e08-5b26-4617-aadd-44b1aac43f12-image.png

          A 1 Reply Last reply Reply Quote 0
          • A Offline
            Affonso @nikade
            last edited by

            @nikade it shows like this

            Screenshot 2024-10-10 at 15.05.43.png

            should I make any change?

            1 Reply Last reply Reply Quote 0
            • olivierlambertO Offline
              olivierlambert Vates 🪐 Co-Founder CEO
              last edited by

              Can you try to add a bit more RAM in case? Like 4GiB/4GiB to see if it's better

              A 1 Reply Last reply Reply Quote 0
              • A Offline
                Affonso @olivierlambert
                last edited by

                @olivierlambert I changed the memory settings like thisScreenshot 2024-10-10 at 16.20.31.png

                I'm going to proceed with the upgrade and post the outcome

                1 Reply Last reply Reply Quote 1
                • olivierlambertO Offline
                  olivierlambert Vates 🪐 Co-Founder CEO
                  last edited by

                  Perfect, keep us posted!

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

                    The same error

                    Screenshot 2024-10-10 at 17.13.40.png

                    1 Reply Last reply Reply Quote 0
                    • olivierlambertO Offline
                      olivierlambert Vates 🪐 Co-Founder CEO
                      last edited by

                      Is there a way we can test that ISO? It's hard to reproduce if we don't have any way to test it here or in the community 😕

                      A 1 Reply Last reply Reply Quote 0
                      • A Offline
                        Affonso @olivierlambert
                        last edited by

                        @olivierlambert let me talk to Netgate support and work some way out

                        1 Reply Last reply Reply Quote 0
                        • nikadeN Offline
                          nikade Top contributor
                          last edited by

                          Googling the error "supervisor read data page not present" gives a lot of hints towards bad memory, are you using ECC or non ECC?

                          A 1 Reply Last reply Reply Quote 0
                          • A Offline
                            Affonso @nikade
                            last edited by

                            @nikade

                            The Dell R620 (Intel):

                            # dmidecode -t memory | grep -i "ecc"
                            	Error Correction Type: Multi-bit ECC
                            

                            The lab AMD Razor

                            # dmidecode -t memory | grep -i "ecc"
                            # 
                            

                            One server has ECC memory, the other doesn't. I believe it would be a very odd case of having "bad memory" on two completely different instances, and the fault being exactly the same on the different machines ... but as I said, this is not my area of expertise.

                            1 Reply Last reply Reply Quote 0
                            • A Online
                              Andrew Top contributor @Affonso
                              last edited by

                              @Affonso I had a problem upgrading OPNsense to 24.7 which uses FreeBSD 14 using XCP 8.2.1

                              OPNsense had a kernel crash related to Xen using some FreeBSD kernel options. OPNsense was able to update their kernel to resolve the crash issue.

                              It was a OPNsense/FreeBSD issue, not a XCP issue. It has been resolved.

                              Here's the OPNsense github issue. I don't know if it's related.

                              A 1 Reply Last reply Reply Quote 1
                              • A Offline
                                Affonso @Andrew
                                last edited by

                                Thank you @Andrew I will mention this to pfSense. lets hope

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

                                  So just to give a quick update on how this went:

                                  Since pfSense 24.03 is based on FreeBSD 15 I proceeded with a FreeBSD 15 installation on XCP-ng to see if the issue stemmed from there. FreeBSD15 installed and booted correctly.
                                  From there I ended up testing the development snapshot 24.08.
                                  Also installed correctly and booted.

                                  So whatever issue was there between pfSense 24.03 and XCP-ng that was preventing it from booting, was only present on version 24.03.

                                  A 1 Reply Last reply Reply Quote 1
                                  • A Online
                                    Andrew Top contributor @Affonso
                                    last edited by

                                    @Affonso Looks like it might have been Bug 15684 in 24.03 that was resolved for 24.11 (release notes).

                                    1 Reply Last reply Reply Quote 2
                                    • First post
                                      Last post