linux - 通过 NFS : Loops on "Sending DHCP and RARP requests ..." 使用根文件系统引导 Linux

Question

我正在尝试通过 NFS 启动 Beaglebone Black，内核仍在我的主机上。我的印象是我的主机没有必要成为 DHCP 服务器，手动 IPv4 静态 IP 就足够了。引导过程最终循环执行以下操作:

[  357.280386] IP-Config: Retrying forever (NFS root)...
[  357.285865] net eth0: initializing cpsw version 1.12 (0)
[  357.393659] SMSC LAN8710/LAN8720 4a101000.mdio:00: attached PHY driver 
[SMSC LAN8710/LAN8720] (mii_bus:phy_addr=4a101000.mdio:00, irq=-1)
[  357.412716] IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready
[  360.563747] cpsw 4a100000.ethernet eth0: Link is Up - 100Mbps/Full - flow control rx/tx
[  360.572294] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[  360.593014] Sending DHCP and RARP requests .....

相关的 bootargs/bootcmd:

U-Boot# echo $nfs_bootcmd                                                      
tftp 0x81000000 zImage; tftp 0x82000000 am335x-boneblack.dtb; bootz 0x81000000 - 0x82000000

U-Boot# echo $bootargs   
root=/dev/nfs rw ip=192.168.0.100:192.168.0.1:::eth0:off console=ttyO0 nfsroot=192.168.0.1:/nfsroot nfsrootdebug

U-Boot# echo $ipaddr     
192.168.0.100

U-Boot# echo $serverip   
192.168.0.1

我的/etc/exports:

/nfsroot *(rw,no_root_squash,no_subtree_check)

这是启动期间的串行输出:Linux boot via NFS .

硬件/软件设置:

• 比格骨黑
• 以太网从主机直接连接到目标板(中间没有路由器/交换机/集线器)
• tftp 工作正常
• rootfs 似乎可以通过 NFS 挂载
• 主机未设置为 dhcp 服务器
• u-boot/Linux

有必要让我的主机成为一个dhcp服务器吗？我该如何解决这个问题？

提前致谢。

Answer 1

我通过在我的主机上设置一个 DHCP 服务器来解决这个问题。安装 isc-dhcp-server(在 Ubuntu 上)后，我将/etc/dhcp/dhcpd.conf 修改为如下所示:

# /etc/dhcp/dhcpd.conf

subnet 192.168.0.0 netmask 255.255.255.0 {
    range 192.168.0.0 192.168.0.254;
}

# the host name doesn't need to match anything.
host beaglebone {
    # MAC address from U-boot's ethaddr env var
    hardware ethernet 04:a3:16:f5:1a:56;

    # The static IP to be allocated to the beaglebone device
    fixed-address 192.168.0.100;
}

这解决了问题并且设备可以正确启动。