Dear List
This thread is still unsolved, any help would be very much appreciated.
On 7/8/13 11:45 AM, Adrian Zaugg wrote:
> What means "re3: watchdog timeout"? Is it possible that the watchdog
> resets the interface?
Please see the full message in the attachment.
Regards, Adrian.
--- Begin Message ---
Thank you for your replies - although they don't satisfy me completely:
Whatever slow hardware I may have, it should work steady, but maybe just
slower. And in my opinion the slave should take over completely not just
the LAN interface, but that's another discussion.
Since my hardware is somewhat reasonable (Dual Via Nano X2 1.2GHz, quad
Realtek RT 8111E on board using driver "re" in an embedded system like
this it isn't an option to change the nic, although I agree that other
brands like intel are preferable to Realtek), I inspected more closely
what happens and found out that the LAN interface on the master is going
down for about 3 seconds (I saw this in the switch's log). The syslog on
the pfsense master shows:
Jul 8 11:27:32 yoghurt check_reload_status: Linkup starting re3
Jul 8 11:27:32 yoghurt kernel: re3: watchdog timeout
Jul 8 11:27:32 yoghurt kernel: re3: link state changed to DOWN
Jul 8 11:27:32 yoghurt kernel: lan_vip1: link state changed to DOWN
Jul 8 11:27:35 yoghurt php: : DEVD Ethernet detached event for lan
Jul 8 11:27:37 yoghurt kernel: lan_vip1: INIT -> BACKUP
Jul 8 11:27:37 yoghurt kernel: re3: link state changed to UP
Jul 8 11:27:37 yoghurt kernel: lan_vip1: link state changed to DOWN
Jul 8 11:27:40 yoghurt kernel: lan_vip1: link state changed to UP
Jul 8 11:27:55 yoghurt php: : Could not send the message to
[email protected] -- Error: could not resolve host "smtp.3eck.net"
Jul 8 11:27:55 yoghurt check_reload_status: Linkup starting re3
Jul 8 11:27:58 yoghurt php: : DEVD Ethernet attached event for lan
Jul 8 11:27:58 yoghurt php: : HOTPLUG: Configuring interface lan
Jul 8 11:27:58 yoghurt kernel: lan_vip1: link state changed to DOWN
Jul 8 11:27:58 yoghurt kernel: lan_vip1: INIT -> BACKUP
Jul 8 11:27:58 yoghurt kernel: lan_vip1: link state changed to DOWN
Jul 8 11:27:59 yoghurt php: : waiting for pfsync...
Jul 8 11:27:59 yoghurt php: : pfsync done in 0 seconds.
Jul 8 11:27:59 yoghurt php: : Configuring CARP settings finalize...
Jul 8 11:28:01 yoghurt kernel: lan_vip1: link state changed to UP
Jul 8 11:28:06 yoghurt check_reload_status: updating dyndns lan
Jul 8 11:28:06 yoghurt php: : Message sent to [email protected] OK
Jul 8 11:28:12 yoghurt kernel: arp: 192.168.254.150 moved from
10:40:f3:f0:67:42 to 78:ca:39:ff:5c:bd on re3
Jul 8 11:28:12 yoghurt php: : Message sent to [email protected] OK
Jul 8 11:28:18 yoghurt php: : Message sent to [email protected] OK
"re3" is the LAN interface.
So I retried with device polling switched on (all hardware off load
option disabled anyway), but it gave me the same results.
Repeating the test from my computer to the pfsense box itself instead
passing the traffic through it, I get a steady stream of 50-55MB/s. No
interruption, no error.
So I conclude that I do not understand the problem: If I get a steady
stream with no interruption to my pfsense box, the hardware and the NIC
driver are working ok. But I do not see a interrelation between an
interface link going down and firewall or CARP settings, they play on
different OSI levels. Could anyone enlighten the situation and solve my
knot?
What means "re3: watchdog timeout"? Is it possible that the watchdog
resets the interface?
Thank you very much for your help.
Best regards, Adrian.
_______________________________________________
List mailing list
[email protected]
http://lists.pfsense.org/mailman/listinfo/list
--- End Message ---
_______________________________________________
List mailing list
[email protected]
http://lists.pfsense.org/mailman/listinfo/list
