High Latency on LAN after power failure - fixed, but not sure how



  • My setup is: Cable Modem > SG-3100 pfSense > HP 1920-24G, then everything connected to HP switch.

    I had a power failure yesterday and, although I have SG-3100 connected to UPS and NUT to shut down, the SG-3100 had lost connectivity to UPS and lost power.

    When I restored power to the full system, I had extreme latency on the LAN, such that connecting to pfSense or HP GUI was very slow, or would just time out.

    I had no idea where to start, but using this guide, https://doc.pfsense.org/index.php/Connectivity_Troubleshooting, I didn't see any problems with WAN Interface or LAN Interface.  The firewall rules did show a large number of entries for Block IPv4 link-logout (1000106631), with a source 169.254.124.103:22936 and destination 255.255.255.255:53273 (port numbers not all the same).

    So, in case there was an issue with a device on LAN (even though there weren't any prior to power failure) I power cycled (couldn't access GUI, probably should have tried CLI) the SG-3100 and connected PC directly to LAN port on SG-3100, removing the HP switch - no improvement.

    Next, I disconnected WAN cycled the SG-3100 and attached LAN (only).  There was temporary improvement and I was able to access more of GUI.  I checked NAT and Firewall rules, and everything was as it had been before, when working fine.  After a few minutes, the system started having the latency again.  I was able to save the configuration (I just did that to test) and I was able to reboot system from GUI.

    Finally, the system started working normally, I reconnected WAN and had normal connectivity and, finally, reconnected the HP switch - all seems to be working fine now.

    Here are my questions: (1) any idea what was happening to cause the latency and (2) how did my steps fix the problem?

    Maybe it's not possible to answer the questions without more detail, but it would be nice to know what's going on - for my own education and so that I can address more quickly if it happens again. Thanks!



  • 169.254… is Windows default IP, this signals DHCP server not operational.  255.255.255.255 is broadcast.  Feels like to me clients are bumping around trying to re-sync, and EVENTUALLY they were successful.  Ur next question, how to configure so things come up quicker next time... wish I knew. :)



  • ^^^^
    The 169.254.0.0 /16 addresses are used on Linux & Macs too.  As you say, they're used when a DHCP server is not available.



  • Thanks for the reply guys.

    That's helpful to know that it was a DHCP server issue.  I could tell that the IP address was indicating something specific, but didn't know what it was.

    Sounds like renewing the DHCP of the WAN (Release, Renew on Status / Interfaces) as described here: https://forum.pfsense.org/index.php?topic=144389.0  might be helpful if it happens again.  Hopefully the UPS/NUT connection will work if we have another outage.



  • 169.254.0.0 /16 is called APIPA (automatic private IP addressing) and not specific to an operating system.
    https://wiki.wireshark.org/APIPA


Log in to reply