sun250无比奇怪的问题sunW hme0 link down,cable problem ?

wangyl1977

前些天有台sun250出问题了
http://bbs.chinaunix.net/viewthr ... &extra=page%3D1,
到现在还没弄好!

用笔记本终于进了系统,除了报c0t0d0s7没mount外,可以正常进系统了!
solaris 8
#ifconfig -a
lo0: flags=849<UP,LOOPBACK,RUNNING,MULTICAST> mtu 8232
        inet 127.0.0.1 netmask ff000000
hme0: flags=863<UP,BROADCAST,NOTRAILERS,RUNNING,MULTICAST> mtu 1500
        inet 1.1.203.195 netmask ffffffc0 broadcast 1.1.203.255
ether 8:0:20:cf:c6:a5
#netstat -nr
Routing Table:
  Destination           Gateway           Flags  Ref   Use   Interface
-------------------- -------------------- ----- ----- ------ ---------
1.1.203.192           1.1.203.195          U        3     151  hme0
224.0.0.0             1.1.203.195          U        3      0  hme0
default               1.1.203.254          UG       0     40  
127.0.0.1             127.0.0.1             UH       0 167431  lo0
#ping 1.1.203.195
1.1.203.195  is alive
#ping 1.1.203.196
显示no answer from 1.1.203.196
1.1.203.195和1.1.203.196 在同一个网段,也不通!
#ping 网关
不通!
系统提示sunW hme0 down,cable problem ?
试过换网线,换交换机口,都不行!
ifconfig hme0 plumb
再重配ip也是不行!

请问可能是什么原因啊? 有人说可能是双工的问题,可是以前一直好好的啊!

briangao

系统提示sunW hme0 down,cable problem ?

This mesage is pretty reliable. It is either a hardware problem or switch port setting problem.
Did you ask your network admin to check the switch port status? If it is errordisabled, it needs to be re-enabled again.
Was there any speed and duplex mismatch for that port setting on the switch?
When you tried another cable, did you plug in the Ethernet cable directly to hme0 and switch port or you plugged it into a wall jack?
When you tried another switch port, did your network admin configure that port to be in the same VLAN?
Worst comes worst, it is your hme0 that has a problem.

wangyl1977

If there is trouble in hme0,why the ping  hme0 itself is ok and the putput of ifconfig is up ???

#ifconfig -a
lo0: flags=849<UP,LOOPBACK,RUNNING,MULTICAST> mtu 8232
        inet 127.0.0.1 netmask ff000000
hme0: flags=863<UP,BROADCAST,NOTRAILERS,RUNNING,MULTICAST> mtu 1500
        inet 1.1.203.195 netmask ffffffc0 broadcast 1.1.203.255
ether 8:0:20:cf:c6:a5
#ping 1.1.203.195
1.1.203.195  is alive

原帖由 briangao 于 2007-3-23 01:45 发表
This mesage is pretty reliable. It is either a hardware problem or switch port setting problem.
Did you ask your network admin to check the switch port status? If it is errordisabled, it needs to ...

briangao

195 is your own IP address. It is always pingable as long as your hme0 is up. If you unplug the the Ethernet cable from hme0, your hme0 will be still up unless you type ifconfig hme0 down. As a matter of fact, the system will only send out "hme0 down,cable problem ?" message if your hme0 is up. If you turn it down, you will not see such message any more. Your hme0 is up and you can ping it doesn't mean you don't have a link problem.

wangyl1977

刚才又去试了试, 更迷惑了!
ifconfig -a的输出和刚才一样,还是Up.
但用ndd -get /dev/hme link_status
0
ndd -get /dev/hme link_speed
0
ndd -get /dev/hme link_mode
0

显示网卡为down,半双工,10M
这个输出和ifconfig -a的输出哪个更准呢? 按理来说ifconfig -a的输出有Up就表示网卡已UP啊,可ndd -get /dev/hme link_status 竟然说网卡已down

原帖由 briangao 于 2007-3-23 09:51 发表
195 is your own IP address. It is always pingable as long as your hme0 is up. If you unplug the the Ethernet cable from hme0, your hme0 will be still up unless you type ifconfig hme0 down. As a mat ...

nus

原帖由 wangyl1977 于 2007-3-23 10:11 发表
刚才又去试了试, 更迷惑了!
ifconfig -a的输出和刚才一样,还是Up.
但用ndd -get /dev/hme link_status
0
ndd -get /dev/hme link_speed
0
ndd -get /dev/hme link_mode
0

显示网卡为down,半双工,10M
...

XD，排除法吧，用交叉线，自己搞个有点逻辑的验证方案。

wangyl1977

你说的是用交叉线搭一个网络?
将笔节本的IP设为和sun的IP一个网段,再互相Ping ?

原帖由 nus 于 2007-3-23 10:15 发表


XD，排除法吧，用交叉线，自己搞个有点逻辑的验证方案。

briangao

ndd -get /dev/hme link_status
0

This means your link is down. hme0 up or down is through manual operation. You can bring hme0 up even you don't have an ethernet cable connected to hme0

wangyl1977

thanks a lot!
you mean my ethernet card is up (just from the  output from ifconfig -a ?) ,but the link is down .
do you have good idea to solove the problem ?
from the output of ndd,the ethernet card is half  duplex ,10M, do i need change the configuration in the switch ?  why the server is ok  just before the crashdown ,we have nor change the configuration in the switch ?

原帖由 briangao 于 2007-3-23 10:37 发表
This means your link is down. hme0 up or down is through manual operation. You can bring hme0 up even you don't have an ethernet cable connected to hme0

wangyl1977

顶一下啊!

原帖由 wangyl1977 于 2007-3-23 12:03 发表
thanks a lot!
you mean my ethernet card is up (just from the  output from ifconfig -a ?) ,but the link is down .
do you have good idea to solove the problem ?
from the output of ndd,the ethernet ...