Comment 7 for bug 1391674

I switched my 3 Supermicro X9SCL/X9SCM servers to 3.19.0-031900rc6-generic and the issue is repeating again under heavy I/O (during backups on DRBD-backed xen guests). The issue happens on all the servers.

Feb 1 00:47:55 hostname kernel: [179433.583173] e1000e 0000:00:19.0 eth1: Detected Hardware Unit Hang:
Feb 1 00:47:55 hostname kernel: [179433.583173] TDH <7c>
Feb 1 00:47:55 hostname kernel: [179433.583173] TDT <89>
Feb 1 00:47:55 hostname kernel: [179433.583173] next_to_use <89>
Feb 1 00:47:55 hostname kernel: [179433.583173] next_to_clean <7a>
Feb 1 00:47:55 hostname kernel: [179433.583173] buffer_info[next_to_clean]:
Feb 1 00:47:55 hostname kernel: [179433.583173] time_stamp <102ab4e95>
Feb 1 00:47:55 hostname kernel: [179433.583173] next_to_watch <7c>
Feb 1 00:47:55 hostname kernel: [179433.583173] jiffies <102ab5089>
Feb 1 00:47:55 hostname kernel: [179433.583173] next_to_watch.status <0>
Feb 1 00:47:55 hostname kernel: [179433.583173] MAC Status <40080183>
Feb 1 00:47:55 hostname kernel: [179433.583173] PHY Status <796d>
Feb 1 00:47:55 hostname kernel: [179433.583173] PHY 1000BASE-T Status <3800>
Feb 1 00:47:55 hostname kernel: [179433.583173] PHY Extended Status <3000>
Feb 1 00:47:55 hostname kernel: [179433.583173] PCI Status <10>

>> uname -a
Linux hostname 3.19.0-031900rc6-generic #201501261152 SMP Mon Jan 26 16:53:27 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux