Menu
▾
▴
[E1000-devel] Re-4: e1000e detected hardware unit hang problem
|
From: Lars M. <ma...@la...> - 2013-03-06 09:58:28
|
Hi Tushar, I tried the "build in" driver of the described kernels 3.3.8, 3.4.32 and 3.7.9 but I had no success. I know this because a have seen the error message three or four times on the screen. Once it was logged in the kern.log as You see here: Mar 1 00:53:36 lightning kernel: e1000e 0000:00:19.0 eth0: Detected Hardware Unit Hang: Mar 1 00:53:36 lightning kernel: TDH <a7> Mar 1 00:53:36 lightning kernel: TDT <c6> Mar 1 00:53:36 lightning kernel: next_to_use <c6> Mar 1 00:53:36 lightning kernel: next_to_clean <a5> Mar 1 00:53:36 lightning kernel: buffer_info[next_to_clean]: Mar 1 00:53:36 lightning kernel: time_stamp <5a0f4f> Mar 1 00:53:36 lightning kernel: next_to_watch <a7> Mar 1 00:53:36 lightning kernel: jiffies <5a1199> Mar 1 00:53:36 lightning kernel: next_to_watch.status <0> Mar 1 00:53:36 lightning kernel: MAC Status <40080083> Mar 1 00:53:36 lightning kernel: PHY Status <796d> Mar 1 00:53:36 lightning kernel: PHY 1000BASE-T Status <3c00> Mar 1 00:53:36 lightning kernel: PHY Extended Status <3000> Mar 1 00:53:36 lightning kernel: PCI Status <10> Mar 1 00:53:38 lightning kernel: e1000e 0000:00:19.0 eth0: Detected Hardware Unit Hang: Mar 1 00:53:38 lightning kernel: TDH <a7> Mar 1 00:53:38 lightning kernel: TDT <c6> Mar 1 00:53:38 lightning kernel: next_to_use <c6> Mar 1 00:53:38 lightning kernel: next_to_clean <a5> Mar 1 00:53:38 lightning kernel: buffer_info[next_to_clean]: Mar 1 00:53:38 lightning kernel: time_stamp <5a0f4f> Mar 1 00:53:38 lightning kernel: next_to_watch <a7> Mar 1 00:53:38 lightning kernel: jiffies <5a138d> Mar 1 00:53:38 lightning kernel: next_to_watch.status <0> Mar 1 00:53:38 lightning kernel: MAC Status <40080083> Mar 1 00:53:38 lightning kernel: PHY Status <796d> Mar 1 00:53:38 lightning kernel: PHY 1000BASE-T Status <3c00> Mar 1 00:53:38 lightning kernel: PHY Extended Status <3000> Mar 1 00:53:38 lightning kernel: PCI Status <10> Mar 1 00:53:40 lightning kernel: e1000e 0000:00:19.0 eth0: Detected Hardware Unit Hang: Mar 1 00:53:40 lightning kernel: TDH <a7> Mar 1 00:53:40 lightning kernel: TDT <c6> Mar 1 00:53:40 lightning kernel: next_to_use <c6> Mar 1 00:53:40 lightning kernel: next_to_clean <a5> Mar 1 00:53:40 lightning kernel: buffer_info[next_to_clean]: Mar 1 00:53:40 lightning kernel: time_stamp <5a0f4f> Mar 1 00:53:40 lightning kernel: next_to_watch <a7> Mar 1 00:53:40 lightning kernel: jiffies <5a1581> Mar 1 00:53:40 lightning kernel: next_to_watch.status <0> Mar 1 00:53:40 lightning kernel: MAC Status <40080083> Mar 1 00:53:40 lightning kernel: PHY Status <796d> Mar 1 00:53:40 lightning kernel: PHY 1000BASE-T Status <3c00> Mar 1 00:53:40 lightning kernel: PHY Extended Status <3000> Mar 1 00:53:40 lightning kernel: PCI Status <10> Mar 1 00:53:42 lightning kernel: e1000e 0000:00:19.0 eth0: Detected Hardware Unit Hang: Mar 1 00:53:42 lightning kernel: TDH <a7> Mar 1 00:53:42 lightning kernel: TDT <c6> Mar 1 00:53:42 lightning kernel: next_to_use <c6> Mar 1 00:53:42 lightning kernel: next_to_clean <a5> Mar 1 00:53:42 lightning kernel: buffer_info[next_to_clean]: Mar 1 00:53:42 lightning kernel: time_stamp <5a0f4f> Mar 1 00:53:42 lightning kernel: next_to_watch <a7> Mar 1 00:53:42 lightning kernel: jiffies <5a1775> Mar 1 00:53:42 lightning kernel: next_to_watch.status <0> Mar 1 00:53:42 lightning kernel: MAC Status <40080083> Mar 1 00:53:42 lightning kernel: PHY Status <796d> Mar 1 00:53:42 lightning kernel: PHY 1000BASE-T Status <3c00> Mar 1 00:53:42 lightning kernel: PHY Extended Status <3000> Mar 1 00:53:42 lightning kernel: PCI Status <10> Mar 1 00:53:43 lightning kernel: ------------[ cut here ]------------ Mar 1 00:53:43 lightning kernel: WARNING: at net/sched/sch_generic.c:255 dev_watchdog+0x1da/0x1f0() Mar 1 00:53:43 lightning kernel: Hardware name: S1200BTL Mar 1 00:53:43 lightning kernel: NETDEV WATCHDOG: eth0 (e1000e): transmit queue 0 timed out Mar 1 00:53:43 lightning kernel: Modules linked in: af_packet xt_REDIRECT xt_DSCP xt_dscp xt_statistic xt_CT xt_NFLOG nfnetlink_log nfnetlink ipt_ULOG xt_LOG xt_time xt_connlimit xt_helper xt_realm xt_NFQUEUE xt_tcpmss xt_tcpudp xt_addrtype xt_pkttype iptable_raw xt_TPROXY nf_tproxy_core xt_CLASSIFY xt_mark xt_hashlimit xt_comment ipt_REJECT xt_length xt_connmark xt_owner xt_recent xt_iprange xt_physdev xt_policy iptable_mangle xt_nat xt_multiport xt_conntrack nf_nat_ftp iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack_ftp nf_conntrack iptable_filter ip_tables x_tables ipmi_poweroff ipmi_devintf ipmi_si ipmi_watchdog ipmi_msghandler minix ppp_mppe ppp_generic slhc tun e1000 usb_storage mousedev usbhid dm_mod microcode i2c_i801 i2c_core sr_mod ehci_hcd e1000e(O) firmware_class cdrom usbcore usb_common evdev unix Mar 1 00:53:43 lightning kernel: Pid: 0, comm: swapper/2 Tainted: G O 3.7.9 #2 Mar 1 00:53:43 lightning kernel: Call Trace: Mar 1 00:53:43 lightning kernel: [<c013696d>] warn_slowpath_common+0x6d/0xa0 Mar 1 00:53:43 lightning kernel: [<c046a06a>] ? dev_watchdog+0x1da/0x1f0 Mar 1 00:53:43 lightning kernel: [<c046a06a>] ? dev_watchdog+0x1da/0x1f0 Mar 1 00:53:43 lightning kernel: [<c0136a1e>] warn_slowpath_fmt+0x2e/0x30 Mar 1 00:53:43 lightning kernel: [<c046a06a>] dev_watchdog+0x1da/0x1f0 Mar 1 00:53:43 lightning kernel: [<c0469e90>] ? pfifo_fast_dequeue+0xe0/0xe0 Mar 1 00:53:43 lightning kernel: [<c0142e7d>] call_timer_fn.isra.32+0x1d/0x80 Mar 1 00:53:43 lightning kernel: [<c0143051>] run_timer_softirq+0x171/0x180 Mar 1 00:53:43 lightning kernel: [<c0469e90>] ? pfifo_fast_dequeue+0xe0/0xe0 Mar 1 00:53:43 lightning kernel: [<c013d6c0>] __do_softirq+0x90/0x140 Mar 1 00:53:43 lightning kernel: [<c013d630>] ? __tasklet_schedule+0x60/0x60 Mar 1 00:53:43 lightning kernel: <IRQ> [<c013d875>] ? irq_exit+0x65/0x70 Mar 1 00:53:43 lightning kernel: [<c0127894>] ? smp_apic_timer_interrupt+0x54/0x90 Mar 1 00:53:43 lightning kernel: [<c04eee3d>] ? apic_timer_interrupt+0x2d/0x34 Mar 1 00:53:43 lightning kernel: [<c02e45b6>] ? acpi_idle_enter_bm+0x251/0x286 Mar 1 00:53:43 lightning kernel: [<c042a335>] ? cpuidle_enter+0x15/0x20 Mar 1 00:53:43 lightning kernel: [<c042a8fe>] ? cpuidle_idle_call+0x6e/0xd0 Mar 1 00:53:43 lightning kernel: [<c0111305>] ? cpu_idle+0x55/0xa0 Mar 1 00:53:43 lightning kernel: [<c04e522b>] ? start_secondary+0x19b/0x1a1 Mar 1 00:53:43 lightning kernel: ---[ end trace a49ef186404ae76d ]--- Mar 1 00:53:43 lightning kernel: e1000e 0000:00:19.0 eth0: Reset adapter unexpectedly Greets Lars Original Message processed by david® RE: Re-2: e1000e detected hardware unit hang problem (06-Mrz-2013 0:17) From: Dave, Tushar N To:Lars Maschke Cc:e10...@li... >-----Original Message----- >From: Lars Maschke [mailto:ma...@la...] >Sent: Monday, March 04, 2013 3:11 PM >To: Dave, Tushar N >Cc: e10...@li... >Subject: Re-2: e1000e detected hardware unit hang problem > >Hello Tushar, > >first of all. Thanks for Your quick reply. > >That's the point. I don't know why this occurs. If I have the chance I see >a failure of the e1000e driver on the console. The server is completly >down and I can't logon to get any other information. > >The error isn't logged at dmesg.log or syslog on my debian system. There >is no logging at all after the crash. We at the least need the dmesg from the kernel to start with. How would you know that you're getting detected hardware unit hang message. Is it possible to take a picture or connect serial console to the server to retrieve the message after crash/hang? > >Only a full reset solves the problem. I get this error every two, three or >four days. At the crash time no special cron job is running. It occurs >only an night between 0:00h and 2:30h. >From now on I try to reset networking with the following bash-script every >night and I hope that it's a good idea: >--- >#!/bin/sh >/etc/init.d/networking stop >/sbin/rmmod e1000e >/sbin/modprobe e1000e RxIntDelay=0,0 IntMode=1,1 /etc/init.d/networking start >/sbin/ethtool -K eth0 tso off /sbin/shorewall restart >--- How many e1000e devices you have in the system? >Do You think that the problem can occur when the other Intel "e1000" >driver is also loaded on the machine? I don't think so however if you can, give it a try. And let me know if anything changes! Have you tried the in-kernel e1000e driver? -Tushar To: tus...@in... Cc: e10...@li... |
