This commit fixes a long standing issue with all my servers running Intel Server NICs with 82571/2 chips: http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=commit;h=09357b00255c233705b1cf6d76a8d147340545b8
I use kernel 2.6.32.XX and 3.2.XX and have the issue on both with two different server NICs (both in the affected chipset range). Eksample output from dmesg when the issue appears: [19185.345835] e1000e 0000:01:00.1: eth1: Detected Hardware Unit Hang: [19185.345837] TDH <5c> [19185.345838] TDT <5d> [19185.345839] next_to_use <5d> [19185.345839] next_to_clean <5c> [19185.345840] buffer_info[next_to_clean]: [19185.345841] time_stamp <1004809ab> [19185.345842] next_to_watch <5c> [19185.345843] jiffies <100480ab8> [19185.345844] next_to_watch.status <0> [19185.345845] MAC Status <80387> [19185.345845] PHY Status <792d> [19185.345846] PHY 1000BASE-T Status <3800> [19185.345847] PHY Extended Status <3000> [19185.345848] PCI Status <10> I tried running kernel 3.3 rc-6 a while just to verify and the issue dissappeared. I would be very happy if this fix could at least be backported to the 3.2.XX kernel, as this is the kernel selected for the next debian release so it will be with us for a long time. Best regards, Hans-Kristian Bakke -- To unsubscribe from this list: send the line "unsubscribe stable" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
