Netgate Discussion Forum
    • Categories
    • Recent
    • Tags
    • Popular
    • Users
    • Search
    • Register
    • Login
    Introducing Netgate Nexus: Multi-Instance Management at Your Fingertips.

    [SOLVED] CPU Load High on V2.8.1

    Scheduled Pinned Locked Moved General pfSense Questions
    16 Posts 4 Posters 669 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.
    • B Offline
      Bambos
      last edited by Bambos

      Hello everyone,

      my pfSense device got unexpectedly high load on CPU out of nowhere.
      it does fluctuate from 35% to 75% , but mostly is around 50%.

      The device is i5-7200U with 16GB Ram , with below settings of crypto.

      bf0202ae-ae38-48e0-bff7-d79998555d1d-image.png

      Does the cryptographic settings affecting this ? (probably)
      what is the case if i select none ?

      530f4e47-1d96-420f-8685-07d6a98914cf-image.png

      Is this normal ? what i have to check ?

      SteveITSS 1 Reply Last reply Reply Quote 0
      • SteveITSS Offline
        SteveITS Rebel Alliance @Bambos
        last edited by

        @Bambos Check Diagnostics > System Activity to see what's using CPU. Or top -aSH via SSH.

        To upgrade, select your branch in System/Update/Update Settings. When upgrading, allow 10-15 minutes to reboot, or more depending on packages, CPU, and/or disk speed.
        Only install packages for your version of pfSense.
        Upvote ๐Ÿ‘ helpful posts!

        B 1 Reply Last reply Reply Quote 0
        • B Offline
          Bambos @SteveITS
          last edited by

          @SteveITS Hello Sir,

          This openvpn server seems strange to me having all this cpu usage.
          How i can identify which one of the openvpn servers is that ?

          (the openvpn servers i can see from services are total 8)

          8c6735bd-2280-4ae8-939c-95792fc547c3-image.png

          0a26c082-f1fe-4001-b769-dd9132568259-image.png

          tinfoilmattT SteveITSS 2 Replies Last reply Reply Quote 0
          • tinfoilmattT Offline
            tinfoilmatt LAYER 8 @Bambos
            last edited by

            When did this start happening?

            B 1 Reply Last reply Reply Quote 0
            • B Offline
              Bambos @tinfoilmatt
              last edited by

              @tinfoilmatt after a total crush of the specific pfsense, no web, no ping, no ssh. no dhcp server, no dns resolve. the only way to revive was a power cycle.

              tinfoilmattT 1 Reply Last reply Reply Quote 0
              • tinfoilmattT Offline
                tinfoilmatt LAYER 8 @Bambos
                last edited by

                When did that crash occur?

                What kind of traffic/user volume do these OVPN servers see? Are they all 'peer-to-peer' (i.e., site-to-site) tunnel? Or is there one (or more) for remote access?

                B 1 Reply Last reply Reply Quote 0
                • B Offline
                  Bambos @tinfoilmatt
                  last edited by Bambos

                  @tinfoilmatt the crush occur last night, when the system was doing nothing. also no change on the configuration last month.

                  the openvpn servers are mixed at this point.
                  2 are remote access user auth
                  1 is remote access SSL/TLS
                  3 are peer to peer SSL/TLS
                  3 are peer to peer shared key

                  users - not much. 1 to 3 users at the same time.
                  peer instances again not much, only 1 LAN for remote access purposes.
                  only 1 instance has some noticeable traffic for offsite backups for some hours once a week.
                  The above config is running for over 4 years from versions 2.5.0 , 2.5.1, 2.5.2 , 2.6.0 , 2.7.1, 2.7.2 , 2.8.1.

                  tinfoilmattT 1 Reply Last reply Reply Quote 0
                  • SteveITSS Offline
                    SteveITS Rebel Alliance @Bambos
                    last edited by

                    @Bambos said in CPU Load High on V2.8.1:

                    How i can identify which one of the openvpn servers is that ?

                    Can you stop them one at a time?

                    To upgrade, select your branch in System/Update/Update Settings. When upgrading, allow 10-15 minutes to reboot, or more depending on packages, CPU, and/or disk speed.
                    Only install packages for your version of pfSense.
                    Upvote ๐Ÿ‘ helpful posts!

                    1 Reply Last reply Reply Quote 1
                    • tinfoilmattT Offline
                      tinfoilmatt LAYER 8 @Bambos
                      last edited by tinfoilmatt

                      Have you checked for any suspicious connection-attempt activity?

                      I'm asking because, as of today, 11 May 2026, CVE-2026-40215 is not patched for pfSense CE yet (and 2.8.1 runs a vulnerable OpenVPN version, 2.6.16. You can run pkg info openvpn to see package version information).

                      B 1 Reply Last reply Reply Quote 0
                      • B Offline
                        Bambos @tinfoilmatt
                        last edited by

                        @tinfoilmatt i confirm that pfSense 2.8.1 and OpenVPN 2.6.16

                        @SteveITS yes i find it. Is a site to site tunnel transferring offsite backups with 4.5MBytes/sec.

                        is that reasonable ? one pfSense using 65% CPU and other 45% CPU for this transfer over VPN.
                        This configuration is shared key. (didn't update / upgrade to SSL/TLS). Is this affecting the CPU Load ?

                        SteveITSS 1 Reply Last reply Reply Quote 0
                        • SteveITSS Offline
                          SteveITS Rebel Alliance @Bambos
                          last edited by

                          @Bambos said in CPU Load High on V2.8.1:

                          is that reasonable ? one pfSense using 65% CPU and other 45% CPU

                          What CPU models are they?

                          What encryption algorithms are used?

                          I don't know the answer, but a slower CPU will use more CPU time for the same amount of calculations.

                          Shared Key has a warning on https://docs.netgate.com/pfsense/en/latest/recipes/openvpn-s2s-psk.html that it is deprecated.

                          To upgrade, select your branch in System/Update/Update Settings. When upgrading, allow 10-15 minutes to reboot, or more depending on packages, CPU, and/or disk speed.
                          Only install packages for your version of pfSense.
                          Upvote ๐Ÿ‘ helpful posts!

                          B 1 Reply Last reply Reply Quote 0
                          • B Offline
                            Bambos @SteveITS
                            last edited by

                            @SteveITS the vpn setup is :
                            Mode: Peer to Peer ( Shared Key )
                            Data Ciphers: AES-128-CBC, AES-128-GCM, AES-256-GCM, AES-256-CBC
                            Digest: SHA256

                            The data rate is 4,5-4,7 MBytes/sec.
                            pfSense 1 has i5-7200U with 16GB Ram (9% used)
                            pfSense 2 has Intel(R) Celeron(R) CPU J1900 with 4GB Ram (12% used)

                            1 Reply Last reply Reply Quote 0
                            • stephenw10S Offline
                              stephenw10 Netgate Administrator
                              last edited by

                              Any OpenVPN traffic will generate CPU load. If the actual bandwidth being passed hasn't changed though then I wouldn't expect to see the CPU usage change. Unless the crypto algorithm has changed. If you control both ends of that tunnel then can just set one cipher so it only ever uses that.

                              tinfoilmattT 1 Reply Last reply Reply Quote 1
                              • tinfoilmattT Offline
                                tinfoilmatt LAYER 8 @stephenw10
                                last edited by

                                OP has yet to confirm if the offsite backup job utilizing the transport tunnel is new or longstanding. But it would seem to be the obvious culprit, so it must be new (and/or jacked up)?

                                1 Reply Last reply Reply Quote 1
                                • stephenw10S Offline
                                  stephenw10 Netgate Administrator
                                  last edited by

                                  Yeah I would first suspect that the backup just got larger and is now taking long enough that it has been noticed.

                                  B 1 Reply Last reply Reply Quote 0
                                  • B Offline
                                    Bambos @stephenw10
                                    last edited by

                                    @stephenw10 yes, i confirm. long backup because of power failure on remote side.
                                    Thanks for all the comments.

                                    1 Reply Last reply Reply Quote 1
                                    • First post
                                      Last post
                                    Copyright 2026 Rubicon Communications LLC (Netgate). All rights reserved.
                                    Privacy Policy · Cookie Policy