Problemas com a placa de rede

brunobonato
(usa Outra)

Enviado em 11/09/2013 - 15:30h

Boa tarde

Tenho um servidor Debian GNU/Linux 6.0 com o kernel na versão 2.6.32-5-686.

Minha placa de rede interna eth1 está caindo sozinha e não volta. Recebi a seguinte mensagem do sistema:


Kernel failure message 1:
NETDEV WATCHDOG: eth1 (e1000e): transmit queue 0 timed out
Modules linked in: acpi_cpufreq cpufreq_powersave cpufreq_userspace cpufreq_stats cpufreq_conservative xt_limit xt_recent xt_state xt_tcpudp xt_multiport nf_nat_ftp nf_conntrack_ftp ipt_MASQUERADE ipt_REDIRECT ipt_LOG iptable_mangle iptable_nat nf_nat nf_conntrack_ipv4 nf_conntrack nf_defrag_ipv4 iptable_filter ip_tables x_tables parport_pc ppdev lp parport sco bridge stp bnep rfcomm l2cap bluetooth rfkill tun nfsd lockd nfs_acl auth_rpcgss sunrpc exportfs binfmt_misc fuse ext3 jbd loop button pcspkr processor serio_raw evdev ext4 mbcache jbd2 crc16 usbhid hid sg sr_mod cdrom hpsa uhci_hcd ata_generic cciss ata_piix tg3 ehci_hcd libphy libata scsi_mod usbcore thermal thermal_sys nls_base e1000e [last unloaded: scsi_wait_scan]
Pid: 0, comm: swapper Not tainted 2.6.32-5-686 #1
Call Trace:
[<c1030949>] ? warn_slowpath_common+0x5e/0x8a
[<c11e9af4>] ? dev_watchdog+0x0/0x15d
[<c10309a7>] ? warn_slowpath_fmt+0x26/0x2a
[<c11e9bb1>] ? dev_watchdog+0xbd/0x15d
[<c1006530>] ? dma_supported+0x63/0x9b
[<c104179a>] ? insert_work+0x71/0x78
[<c1041b13>] ? delayed_work_timer_fn+0x0/0x28
[<c103b208>] ? run_timer_softirq+0x16a/0x1eb
[<c1035a96>] ? __do_softirq+0xaa/0x156
[<c1035b73>] ? do_softirq+0x31/0x3c
[<c1035c4d>] ? irq_exit+0x26/0x58
[<c10144d5>] ? smp_apic_timer_interrupt+0x6c/0x76
[<c1003b35>] ? apic_timer_interrupt+0x31/0x38
[<c10088fb>] ? mwait_idle+0x62/0x6c
[<c1002377>] ? cpu_idle+0x89/0xa2
---[ end trace 08de60c183a164f9 ]---


Kernel failure message 2:
------------[ cut here ]------------
WARNING: at /build/buildd-linux-2.6_2.6.32-45-i386-yQfQSv/linux-2.6-2.6.32/debian/build/source_i386_none/net/sched/sch_generic.c:261 dev_watchdog+0xbd/0x15d()
Hardware name: ProLiant ML150 G6
NETDEV WATCHDOG: eth1 (e1000e): transmit queue 0 timed out
Modules linked in: acpi_cpufreq cpufreq_powersave cpufreq_userspace cpufreq_stats cpufreq_conservative xt_limit xt_recent xt_state xt_tcpudp xt_multiport nf_nat_ftp nf_conntrack_ftp ipt_MASQUERADE ipt_REDIRECT ipt_LOG iptable_mangle iptable_nat nf_nat nf_conntrack_ipv4 nf_conntrack nf_defrag_ipv4 iptable_filter ip_tables x_tables parport_pc ppdev lp parport sco bridge stp bnep rfcomm l2cap bluetooth rfkill tun nfsd lockd nfs_acl auth_rpcgss sunrpc exportfs binfmt_misc fuse ext3 jbd loop button pcspkr processor serio_raw evdev ext4 mbcache jbd2 crc16 usbhid hid sg sr_mod cdrom hpsa uhci_hcd ata_generic cciss ata_piix tg3 ehci_hcd libphy libata scsi_mod usbcore thermal thermal_sys nls_base e1000e [last unloaded: scsi_wait_scan]
Pid: 0, comm: swapper Not tainted 2.6.32-5-686 #1
Call Trace:
[<c1030949>] ? warn_slowpath_common+0x5e/0x8a
[<c11e9af4>] ? dev_watchdog+0x0/0x15d
[<c10309a7>] ? warn_slowpath_fmt+0x26/0x2a
[<c11e9bb1>] ? dev_watchdog+0xbd/0x15d
[<c1006530>] ? dma_supported+0x63/0x9b
[<c104179a>] ? insert_work+0x71/0x78
[<c1041b13>] ? delayed_work_timer_fn+0x0/0x28
[<c103b208>] ? run_timer_softirq+0x16a/0x1eb
[<c1035a96>] ? __do_softirq+0xaa/0x156
[<c1035b73>] ? do_softirq+0x31/0x3c
[<c1035c4d>] ? irq_exit+0x26/0x58
[<c10144d5>] ? smp_apic_timer_interrupt+0x6c/0x76
[<c1003b35>] ? apic_timer_interrupt+0x31/0x38
[<c10088fb>] ? mwait_idle+0x62/0x6c
[<c1002377>] ? cpu_idle+0x89/0xa2
---[ end trace 08de60c183a164f9 ]---



Alguém consegue me dizer o que quer dizer isso?
É minha placa de rede?
É o kernel que está indo pro pau?

O que será que está acontecendo.

Se eu colocar ifdown eth1 e ifup eth1 ele volta a funcionar. Só que ta caindo de 1 a 2 vezes por dia.

A placa de rede é uma Broadcom Corporation NetXtreme BCM5723 Gigabit Ethernet PCIe.

Agradeço desde já.