8.1-RC1: re0: watchdog timeout

I've never seen this symptom before upgrading to 8.1-RC1. But every once in a while (averaging about once an hour), the Realtek gigabit Ethernet interface on my system's motherboard does this:

Code:
re0: watchdog timeout
re0: link state changed to DOWN
re0: link state changed to UP

During which time the system is unresponsive (from the network). But then it picks up as usual.

I've got it set up for DEVICE_POLLING - but I had it set up that way with 8.0-RELEASE and it worked fine.

Anyone know what's going on?
 
Same trouble using 8.1-RELEASE (8-STABLE as well):
Code:
Aug  3 13:46:27 darktower kernel: bge0: watchdog timeout -- resetting
Aug  3 13:46:27 darktower kernel: bge0: link state changed to DOWN
Aug  3 13:46:29 darktower kernel: bge0: link state changed to UP
Aug  3 13:47:54 darktower kernel: bge0: watchdog timeout -- resetting
Aug  3 13:47:54 darktower kernel: bge0: link state changed to DOWN
Aug  3 13:47:56 darktower kernel: bge0: link state changed to UP
Aug  3 13:49:03 darktower kernel: bge0: watchdog timeout -- resetting
Aug  3 13:49:03 darktower kernel: bge0: link state changed to DOWN
Aug  3 13:49:05 darktower kernel: bge0: link state changed to UP
Aug  3 13:50:09 darktower kernel: bge0: watchdog timeout -- resetting
Aug  3 13:50:09 darktower kernel: bge0: link state changed to DOWN
Aug  3 13:50:11 darktower kernel: bge0: link state changed to UP

I'm connected to Internet using DHCP. This problem kills my connections - I loose IP lease and should restart dhclient to accuire new one.
 
knarf said:

If I read that change correctly, it says to stop applying that restriction to a specific variant covered by the driver, but in my case, my kernel says

Code:
re0: <RealTek 8168/8111 B/C/CP/D/DP/E PCIe Gigabit Ethernet> port 0xb800-0xb8ff mem 0xfe9ff000-0xfe9fffff irq 19 at device 0.0 on pci3

Which, by my calculation, isn't the kind excluded by that patch.

In other words, that change is a no-op for me, so far as I can tell.
 
Back
Top