2.3 Wake on LAN



  • Hi.  I upgraded to 2.3 today and the two things I have noticed is my Wake on LAN no longer works.  I used WOL almost every day on the previous version and it would send the magic packet to my PC not that does not seem to be happening.  I am not able to wake on lan from command line either (No, there have been no configuration changes to my PC that would break wake on lan)

    The other thing I notice is that my gateway status always says down and even after changing the monitoring IP it still shows as down.  Something wrong with dpinger?

    WOL is very important to me.  Is there a way I can roll back to the previous version without having to reload from scratch?


  • Rebel Alliance Developer Netgate

    Does the magic packet not actually leave your system? Can you check with a packet capture?



  • Looks like bad cksum. Here is the packet capture after sending hte magic packet.

    14:40:52.657088 IP (tos 0x0, ttl 64, id 1494, offset 0, flags [none], proto UDP (17), length 130, bad cksum 0 (->4ae2)!)
        10.90.10.1.32343 > 10.90.10.255.40000: UDP, length 102
    14:40:52.657146 IP (tos 0x0, ttl 64, id 1494, offset 0, flags [none], proto UDP (17), length 130, bad cksum 0 (->4ae2)!)
        10.90.10.1.32343 > 10.90.10.255.40000: UDP, length 102


  • Rebel Alliance Developer Netgate

    No that's not a bad checksum, it's 0 meaning the hardware (NIC in this case) offloaded the checksum.

    Crank up the detail on that, check the destination MAC.



  • Here is the full detail level.  I know the MAC is right because it's been working for months with no issues.  Tonight I will power on my PC then shut it down and see if I can wake it up then but it has worked flawlessly until I updated.  Is there a log I can tail that will show this activity as it is happening?  Also I did try removing the WoL entry and re-adding it.  I will also mention that nowhere in this packet capture do I see the mac address of my PC that I am trying to wake up

    14:56:10.408825 00:30:18:aa:ca:0e > ff:ff:ff:ff:ff:ff, ethertype IPv4 (0x0800), length 144: (tos 0x0, ttl 64, id 47475, offset 0, flags [none], proto UDP (17), length 130, bad cksum 0 (->9744)!)
        10.90.10.1.22868 > 10.90.10.255.40000: [bad udp cksum 0x2a33 -> 0x8e28!] UDP, length 102
    14:56:10.408845 00:30:18:aa:ca:0e > ff:ff:ff:ff:ff:ff, ethertype IPv4 (0x0800), length 144: (tos 0x0, ttl 64, id 47475, offset 0, flags [none], proto UDP (17), length 130, bad cksum 0 (->9744)!)
        10.90.10.1.22868 > 10.90.10.255.40000: [bad udp cksum 0x2a33 -> 0x8e28!] UDP, length 102
    14:56:24.370709 14:49:e0:09:c3:8b > 00:30:18:aa:ca:0e, ethertype IPv4 (0x0800), length 98: (tos 0x0, ttl 64, id 0, offset 0, flags [DF], proto ICMP (1), length 84)
        10.90.10.20 > 10.90.10.1: ICMP echo request, id 5325, seq 0, length 64
    14:56:24.370773 00:30:18:aa:ca:0e > 14:49:e0:09:c3:8b, ethertype IPv4 (0x0800), length 98: (tos 0x0, ttl 64, id 64945, offset 0, flags [DF], proto ICMP (1), length 84, bad cksum 0 (->142f)!)


  • Rebel Alliance Developer Netgate

    No log or anything. It's really trivial. It just runs a command to send out the WOL magic packet using the MAC you specify, it doesn't get any simpler. I don't see what could have broken.



  • I appreciate your time.  I will update once I have rebooted and everything tonight.  Another thing I notice is my user/certificate no longer show in the openvpn client export section so I am unable to export vpn keys like I used to be able to.  Sorry if this is the wrong place to report these issues.  I just came to the forums today so I could let the community know what I was seeing.


  • Rebel Alliance Developer Netgate

    Start a new thread for a new issue, in the OpenVPN forum for that one.



  • Wake on LAN is working for myself on 2.3.



  • turns out my pc was on configure updates after reboot so apparently that was preventing it from being woke up.  I haven't seen that before so I didn't expect that as an issue.  Thank you for your replies on this.


  • Rebel Alliance Developer Netgate

    Good to know it wasn't a problem – if it was, it would have been quite unusual given it appeared to be working correctly in every way. :-)