0005618: NMI received for unknown reason

We have a just purchased a number of Dell C2100 servers running CentOS 6.2 (2.6.32-220.7.1.el6.x86_64 kernel). After approx 5 mins I get the following error on the console:<br /> <br /> Uhhuh. NMI received for unknown reason 2d on CPU 0.<br /> Do you have a strange power saving mode enabled?<br /> Dazed and confused, but trying to continue.<br /> <br /> And the 4 port gigabit ethernet adaptor goes offline:<br /> idb 0000:06:00.0: eth0 reset adapter<br /> idb 0000:07:00.1: eth3 reset adapter<br /> idb 0000:06:00.1: eth1 reset adapter<br /> idb 0000:07:00.0: eth2 reset adapter<br /> <br /> Output from dmesg:<br /> Uhhuh. NMI received for unknown reason 2d on CPU 0.<br /> Do you have a strange power saving mode enabled?<br /> Dazed and confused, but trying to continue<br /> ------------[ cut here ]------------<br /> WARNING: at net/sched/sch_generic.c:261 dev_watchdog+0x26d/0x280() (Not tainted)<br /> Hardware name: PowerEdge C2100 <br /> NETDEV WATCHDOG: eth3 (igb): transmit queue 0 timed out<br /> Modules linked in: ipmi_si mpt2sas scsi_transport_sas raid_class mptctl mptbase ipmi_devintf ipmi_msghandler dell_rbu 8021q garp stp llc bonding ipv6 dm_mod ses enclosure sg igb dca dcdbas serio_raw i2c_i801 i2c_core iTCO_wdt iTCO_vendor_support i7core_edac edac_core shpchp ext3 jbd mbcache sd_mod crc_t10dif megaraid_sas pata_acpi ata_generic ata_piix [last unloaded: ipmi_si]<br /> Pid: 0, comm: swapper Not tainted 2.6.32-220.7.1.el6.x86_64 <a href="http://bugs.centos.org/view.php?id=1">0000001</a><br /> Call Trace:<br /> <IRQ> [<ffffffff81069a17>] ? warn_slowpath_common+0x87/0xc0<br /> [<ffffffff81069b06>] ? warn_slowpath_fmt+0x46/0x50<br /> [<ffffffff8144a60d>] ? dev_watchdog+0x26d/0x280<br /> [<ffffffff8107cff4>] ? mod_timer+0x144/0x220<br /> [<ffffffff8144a3a0>] ? dev_watchdog+0x0/0x280<br /> [<ffffffff8107c7f7>] ? run_timer_softirq+0x197/0x340<br /> [<ffffffff810a0b20>] ? tick_sched_timer+0x0/0xc0<br /> [<ffffffff8102af2d>] ? lapic_next_event+0x1d/0x30<br /> [<ffffffff81072001>] ? __do_softirq+0xc1/0x1d0<br /> [<ffffffff81095610>] ? hrtimer_interrupt+0x140/0x250<br /> [<ffffffff8100c24c>] ? call_softirq+0x1c/0x30<br /> [<ffffffff8100de85>] ? do_softirq+0x65/0xa0<br /> [<ffffffff81071de5>] ? irq_exit+0x85/0x90<br /> [<ffffffff814f4eb0>] ? smp_apic_timer_interrupt+0x70/0x9b<br /> [<ffffffff8100bc13>] ? apic_timer_interrupt+0x13/0x20<br /> <EOI> [<ffffffff812c4b0e>] ? intel_idle+0xde/0x170<br /> [<ffffffff812c4af1>] ? intel_idle+0xc1/0x170<br /> [<ffffffff813fa027>] ? cpuidle_idle_call+0xa7/0x140<br /> [<ffffffff81009e06>] ? cpu_idle+0xb6/0x110<br /> [<ffffffff814d420a>] ? rest_init+0x7a/0x80<br /> [<ffffffff81c1ff76>] ? start_kernel+0x424/0x430<br /> [<ffffffff81c1f33a>] ? x86_64_start_reservations+0x125/0x129<br /> [<ffffffff81c1f438>] ? x86_64_start_kernel+0xfa/0x109<br /> ---[ end trace 120c4b9c89ff5465 ]---<br /> igb 0000:07:00.1: eth3: Reset adapter<br /> bonding: bond0: link status definitely down for interface eth3, disabling it<br /> igb 0000:06:00.0: eth0: Reset adapter<br /> bonding: bond0: link status definitely down for interface eth0, disabling it<br /> igb 0000:06:00.1: eth1: Reset adapter<br /> bonding: bond0: link status definitely down for interface eth1, disabling it<br /> igb 0000:07:00.0: eth2: Reset adapter<br /> bonding: bond0: link status definitely down for interface eth2, disabling it<br /> <br /> I've got CentOS 5.7 installed on c2100s as well which don't experience this issue.