I an trying to download large amount of data through NFS share. After a short time network link became unusable on all protocols (ssh, ping, NFS, http). It happens on the network card which is used to download large amount of data (I have three cards from different vendors). after ifconfig eth2 down && ifconfig eth2 up network work again for about two minutes. nothing is written in /var/log/messages I use 2.6.29-r1 and willing to make tests on request. On the same machine I have another bug. I don't know are they related or not: [69062.698027] ------------[ cut here ]------------ [69062.698031] WARNING: at net/sched/sch_generic.c:226 dev_watchdog+0xe7/0x148() [69062.698033] Hardware name: EP35-DS3R [69062.698035] NETDEV WATCHDOG: eth0 (r8169): transmit timed out [69062.698037] Modules linked in: cifs r8169 8139too ipg dm_snapshot scsi_wait_s can [last unloaded: i2c_i801] [69062.698046] Pid: 0, comm: swapper Tainted: G W 2.6.29-gentoo-r1 #6 [69062.698048] Call Trace: [69062.698050] <IRQ> [<ffffffff8105761b>] warn_slowpath+0xd3/0xf2 [69062.698058] [<ffffffff814f4e86>] ? br_nf_pre_routing+0x4b2/0x4d4 [69062.698062] [<ffffffff812be22b>] ? cpumask_next+0x19/0x1b [69062.698065] [<ffffffff812be2cd>] ? cpumask_next_and+0x20/0x33 [69062.698070] [<ffffffff8104e46b>] ? find_busiest_group+0x25a/0x784 [69062.698074] [<ffffffff81071035>] ? clocksource_read+0x7/0x9 [69062.698076] [<ffffffff81071672>] ? getnstimeofday+0x5a/0xbb [69062.698080] [<ffffffff8152ccfe>] ? _spin_lock+0x9/0xc [69062.698083] [<ffffffff8144a245>] ? __netif_tx_lock+0x16/0x1f [69062.698086] [<ffffffff8144a375>] ? netif_tx_lock+0x56/0x70 [69062.698089] [<ffffffff8144a3fa>] ? dev_watchdog+0x0/0x148 [69062.698091] [<ffffffff8144a4e1>] dev_watchdog+0xe7/0x148 [69062.698095] [<ffffffff8102bcd9>] ? read_tsc+0xd/0x24 [69062.698098] [<ffffffff810603ea>] ? cascade+0x4c/0x69 [69062.698101] [<ffffffff81060554>] run_timer_softirq+0x14d/0x1d0 [69062.698104] [<ffffffff8105c626>] __do_softirq+0x7e/0x12d [69062.698108] [<ffffffff8102755c>] call_softirq+0x1c/0x28 [69062.698111] [<ffffffff810287ea>] do_softirq+0x44/0x92 [69062.698114] [<ffffffff8105c510>] irq_exit+0x3f/0x79 [69062.698118] [<ffffffff810372ee>] smp_apic_timer_interrupt+0x77/0x84 [69062.698121] [<ffffffff81026f33>] apic_timer_interrupt+0x13/0x20 [69062.698123] <EOI> [<ffffffff81097cba>] ? rcu_pending+0x5d/0x69 [69062.698129] [<ffffffff8102cbb2>] ? mwait_idle+0x60/0x6e [69062.698131] [<ffffffff8102cba5>] ? mwait_idle+0x53/0x6e [69062.698134] [<ffffffff81024b7b>] ? cpu_idle+0x59/0x88 [69062.698138] [<ffffffff81527624>] ? start_secondary+0x191/0x195 [69062.698141] ---[ end trace f33d537881e0b97f ]---
Created attachment 188874 [details] emerge --info
Created attachment 188876 [details] zcat /proc/config.gz
Also, it is good to learn, how to remove "Tainted" mark :)
Could this be related to the upstream bug: http://bugzilla.kernel.org/show_bug.cgi?id=12500 Similar traces and same causes. (nic under load)
Yes, the trace message suggests the same as your bug description and that upstream bug looks correct. Will track that. If you want to help further, please see the task I just suggested on the upstream bug :)
