Hotplug event - drops network of 50 users regularly
-
I am using pfsense 2.0.1 - It has been online for 38 days without this issue. The client called me today to let me know their connection is up an down.
**Update I checked the logs and I am getting about 8 hotplug events nearly everyday between UTC 1600 - 2200. FYI I have LAN port of the pfsense plugged into the Gig port on a Dell switch and I'm using a 10/100 NIC, could this be the issue? Is there anyway to stop this from happening again?
*Unplugged LAN cable from switch and plugged it in to a 10/100 port, still same problemJul 30 16:34:19 php: : Hotplug event detected for lan but ignoring since interface is configured with static IP (10.250.18.1)
Jul 30 16:34:21 php: : Hotplug event detected for lan but ignoring since interface is configured with static IP (10.250.18.1)
Jul 30 16:34:21 check_reload_status: rc.newwanip starting dc0
Jul 30 16:34:26 php: : rc.newwanip: Informational is starting dc0.
Jul 30 16:34:26 php: : rc.newwanip: on (IP address: 10.250.18.1) (interface: lan) (real interface: dc0).
Jul 30 16:34:26 apinger: Exiting on signal 15.
Jul 30 16:34:27 check_reload_status: Reloading filter
Jul 30 16:34:27 apinger: Starting Alarm Pinger, apinger(24184)
Jul 30 16:34:47 dnsmasq[56252]: reading /etc/resolv.conf
Jul 30 16:34:47 dnsmasq[56252]: using nameserver 8.8.4.4#53
Jul 30 16:34:47 dnsmasq[56252]: using nameserver 8.8.8.8#53
Jul 30 16:34:47 dnsmasq[56252]: ignoring nameserver 127.0.0.1 - local interface
Jul 30 16:34:47 dnsmasq[56252]: ignoring nameserver 127.0.0.1 - local interface
Jul 30 16:55:39 check_reload_status: Linkup starting dc0
Jul 30 16:55:39 kernel: dc0: link state changed to DOWN
Jul 30 16:55:42 check_reload_status: Linkup starting dc0
Jul 30 16:55:42 kernel: dc0: link state changed to UP
Jul 30 16:55:44 php: : Hotplug event detected for lan but ignoring since interface is configured with static IP (10.250.18.1)
Jul 30 16:55:48 php: : Hotplug event detected for lan but ignoring since interface is configured with static IP (10.250.18.1)
Jul 30 16:55:48 check_reload_status: rc.newwanip starting dc0
Jul 30 16:55:53 php: : rc.newwanip: Informational is starting dc0.
Jul 30 16:55:53 php: : rc.newwanip: on (IP address: 10.250.18.1) (interface: lan) (real interface: dc0).
Jul 30 16:55:53 apinger: Exiting on signal 15.
Jul 30 16:55:54 check_reload_status: Reloading filter
Jul 30 16:55:54 apinger: Starting Alarm Pinger, apinger(153)
Jul 30 16:56:14 dnsmasq[56252]: reading /etc/resolv.conf
Jul 30 16:56:14 dnsmasq[56252]: using nameserver 8.8.4.4#53
Jul 30 16:56:14 dnsmasq[56252]: using nameserver 8.8.8.8#53
Jul 30 16:56:14 dnsmasq[56252]: ignoring nameserver 127.0.0.1 - local interface
Jul 30 16:56:14 dnsmasq[56252]: ignoring nameserver 127.0.0.1 - local interface
Jul 30 16:56:15 check_reload_status: Linkup starting dc0
Jul 30 16:56:15 kernel: dc0: link state changed to DOWN
Jul 30 16:56:17 check_reload_status: Linkup starting dc0
Jul 30 16:56:17 kernel: dc0: link state changed to UP
Jul 30 16:56:20 php: : Hotplug event detected for lan but ignoring since interface is configured with static IP (10.250.18.1)
Jul 30 16:56:22 php: : Hotplug event detected for lan but ignoring since interface is configured with static IP (10.250.18.1)
Jul 30 16:56:22 check_reload_status: rc.newwanip starting dc0
Jul 30 16:56:27 php: : rc.newwanip: Informational is starting dc0.
Jul 30 16:56:27 php: : rc.newwanip: on (IP address: 10.250.18.1) (interface: lan) (real interface: dc0).
Jul 30 16:56:27 apinger: Exiting on signal 15.
Jul 30 16:56:28 check_reload_status: Reloading filter
Jul 30 16:56:28 apinger: Starting Alarm Pinger, apinger(13524)
Jul 30 16:56:49 dnsmasq[56252]: reading /etc/resolv.conf
Jul 30 16:56:49 dnsmasq[56252]: using nameserver 8.8.4.4#53
Jul 30 16:56:49 dnsmasq[56252]: using nameserver 8.8.8.8#53
Jul 30 16:56:49 dnsmasq[56252]: ignoring nameserver 127.0.0.1 - local interface
Jul 30 16:56:49 dnsmasq[56252]: ignoring nameserver 127.0.0.1 - local interface
Jul 30 17:13:54 check_reload_status: Syncing firewall -
No one has a clue what may be the problem with this system? Need help bad, please! :-[
-
No one has a clue what may be the problem with this system?
It is a nasty problem. It appears device dc0 is reporting loss of carrier from the system (or switch) at the other end of the cable plugged into dc0. Some possible causes:
1. There is a problem with the cable plugged into dc0 or its connection to the socket at either end. Perhaps the cable has been crushed and some of the conductors intermittently go open circuit. Perhaps one or both ends are not correctly seated in the socket. Perhaps the cable is routed in such a way that someone or something intermittently tugs it and causes the plugs to move slightly in the socket.
2. There is a problem with the interface dc0 - try another interface as lan interface and see if the problem moves.
3. There is a problem in the dc driver - try another interface type as lan (e.g. if your physical interfaces are dc0 and re0 switch the roles of dc0 and re0)
4. power supply to dc0 is marginal and intermittently (e.g. under heavy load on the box or heavy load on the mains grid) the power available to dc0 drops below its needs.dc interfaces are now pretty old. It is possible some components on the board have drifted significantly from their nominal values due to age.
-
Thank you sooo much! FYI-It does seem to happen under a more heavy load and time of day.
Here is my plan.
1 - Have the customer replace the network cable to dc0.
2 - If there is still an issue, I built up a new HP Pentium 4 system and added two Intel cards to it and disabled the on-board NIC and disabled PnP. I restored the config to it and will have the customer start using this one.I will let you know the results!