2.6.12-rc2 netconsole problem

2.6.12-rc2 netconsole problem

[Sam Song]

Hi all,

I tried netconsole on a 8241 board with RealTek 8110S 
Gb Ethernet.It seemed nice so far but still one step 
to make it. Did I miss sth or what?

Linux version 2.6.12-rc2 (root@sam.shu.org) (gcc
version 3.2.2 20030217 (Yellow Dog Linux 3.0
3.2.2-2a_1)) #3 Mon Jul 4 12:33:23 CST 2005
......
Kernel command line:
netconsole=@192.168.57.243/eth0,@192.168.57.200/ 
root=/dev/ram0 rw
ip=192.168.57.243:192.168.57.200:::TOPAZ:eth0:off 
panic=1
netconsole: local port 6665
netconsole: local IP 192.168.57.243
netconsole: interface eth0
netconsole: remote port 6666
netconsole: remote IP 192.168.57.200
netconsole: remote ethernet address ff:ff:ff:ff:ff:ff
......
RAMDISK driver initialized: 16 RAM disks of 200000K 
size 1024 blocksize
eth0: RTL8169s/8110s Gigabit Ethernet driver 2.2 at 
0xfe00, 00:d0:b8:00:c3:52, IRQ 0
eth0: Auto-negotiation Enabled.
eth0: 100Mbps Full-duplex operation.
netconsole: device eth0 not up yet, forcing it
netconsole: carrier detect appears untrustworthy,
waiting 4 seconds
netconsole: network logging started
Uniform Multi-Platform E-IDE driver Revision:
7.00alpha2
ide: Assuming 33MHz system bus speed for PIO modes;
override with idebus=xx
......
NET: Registered protocol family 2
IP: routing cache hash table of 512 buckets, 4Kbytes
TCP established hash table entries: 4096 (order: 3,
32768 bytes)
TCP bind hash table entries: 4096 (order: 2, 16384
bytes)
TCP: Hash tables configured (established 4096 bind
4096)
NET: Registered protocol family 1
NET: Registered protocol family 17
IP-Config: Guessing netmask 255.255.255.0
IP-Config: Complete:
      device=eth0, addr=192.168.57.243,
mask=255.255.255.0, gw=255.255.255.255,
     host=192, domain=, nis-domain=(none),
     bootserver=192.168.57.200,
rootserver=192.168.57.200, rootpath=
RAMDISK: Compressed image found at block 0
EXT2-fs warning: mounting unchecked fs, running e2fsck
is recommended
VFS: Mounted root (ext2 filesystem).
Mounted devfs on /dev
Freeing unused kernel memory: 120k init

OK, it just stoped before init process. Is there any
necessary to fix sth on RAMDISK like add netconsole
device node?

Thanks in advance,

Sam


	

	
		
___________________________________________________________ 
雅虎免费G邮箱－中国第一绝无垃圾邮件骚扰超大邮箱
http://cn.mail.yahoo.com/?id=77071

Re: 2.6.12-rc2 netconsole problem

[Tom Rini]

On Tue, Jul 05, 2005 at 09:55:13PM +0800, Sam Song wrote:

[snip]
> OK, it just stoped before init process. Is there any
> necessary to fix sth on RAMDISK like add netconsole
> device node?

Nope, that's how netconsole works, it's not interactive (like console is
after init is spawned).

-- 
Tom Rini
http://gate.crashing.org/~trini/

Re: 2.6.12-rc2 netconsole problem

[Sam Song]

Tom Rini <trini@kernel.crashing.org> wrote:
> [snip]
> > OK, it just stoped before init process. Is there
> > any necessary to fix sth on RAMDISK like add
> > netconsole device node?
> 
> Nope, that's how netconsole works, it's not
> interactive (like console is
> after init is spawned).

Not interactive? I meant to use netconsole to repalce
the mointor serial console. Meanwhile, I implemented 
netconsole on the target in u-boot and it did get a
interactive console. Maybe I can try telnet after 
netconsole to get a net console in Linux? Sound a 
little bit complicated than what I thought:-)

Uh, is this the real reason that netconsole in Linux
_must_ disable ethernet interrupt whereas network 
does need it?

Thanks a lot,

Sam


		
___________________________________________________________ 
雅虎免费G邮箱－No.1的防毒防垃圾超大邮箱 
http://cn.mail.yahoo.com/?id=77072

Re: 2.6.12-rc2 netconsole problem

[Tom Rini]

On Wed, Jul 06, 2005 at 12:10:38PM +0800, Sam Song wrote:
> Tom Rini <trini@kernel.crashing.org> wrote:
> > [snip]
> > > OK, it just stoped before init process. Is there
> > > any necessary to fix sth on RAMDISK like add
> > > netconsole device node?
> > 
> > Nope, that's how netconsole works, it's not
> > interactive (like console is
> > after init is spawned).
> 
> Not interactive? I meant to use netconsole to repalce
> the mointor serial console. Meanwhile, I implemented 
> netconsole on the target in u-boot and it did get a
> interactive console. Maybe I can try telnet after 
> netconsole to get a net console in Linux? Sound a 
> little bit complicated than what I thought:-)

So netconsole is a write-only kernel console.  Kernel messages will go
out, but your init process, etc will not.  For syslog type things, you
may want to look into a syslogd that does network.  I don't know how
netconsole support in u-boot works, but it sounds like it's a network
aware application that you can telnet into, rather than Linux's
netconsole.

-- 
Tom Rini
http://gate.crashing.org/~trini/

Re: 2.6.12-rc2 netconsole problem

[Wolfgang Denk]

In message <20050706041038.70258.qmail@web15810.mail.cnb.yahoo.com> you wrote:
>
> > Nope, that's how netconsole works, it's not
> > interactive (like console is
> > after init is spawned).
> 
> Not interactive? I meant to use netconsole to repalce
> the mointor serial console. Meanwhile, I implemented 
> netconsole on the target in u-boot and it did get a
> interactive console. Maybe I can try telnet after 

It seems you failed to read the doc/README.NetConsole fle that  comes
with  the  U-Boot sources. This explains clearly that while in U-Boot
the netconsole implementation is bidirectional, the  Linux  coede  is
unidirectional only.

> netconsole to get a net console in Linux? Sound a 
> little bit complicated than what I thought:-)

You should not use telnet for  obvious  security  problems.  Use  ssh
instead.  But  from  the technical side it is perfectly fine to use a
remote login through telnet or rsh or ssh...

Best regards,

Wolfgang Denk

-- 
Software Engineering:  Embedded and Realtime Systems,  Embedded Linux
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd@denx.de
                  Nail here --X-- for new monitor.

Re: 2.6.12-rc2 netconsole problem

[Wolfgang Denk]

In message <20050706051139.GB8358@smtp.west.cox.net> you wrote:
>
> may want to look into a syslogd that does network.  I don't know how
> netconsole support in u-boot works, but it sounds like it's a network
> aware application that you can telnet into, rather than Linux's
> netconsole.

It's an implementation which is compatible with  (and  derived  from)
Linux'  netconsole,  but  with  an  added  input  channel so you have
interactive control over ethernet.

Best regards,

Wolfgang Denk

-- 
Software Engineering:  Embedded and Realtime Systems,  Embedded Linux
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd@denx.de
You got to learn three things. What's  real,  what's  not  real,  and
what's the difference."           - Terry Pratchett, _Witches Abroad_

Re: 2.6.12-rc2 netconsole problem

[Sam Song]

Tom Rini <trini@kernel.crashing.org> wrote:
> > > [snip]
> > > Nope, that's how netconsole works, it's not
> > > interactive (like console is
> > > after init is spawned).
> > 
> > Not interactive? I meant to use netconsole to
> > repalce the mointor serial console. Meanwhile, I
> > implemented netconsole on the target in u-boot 
> > and it did get a interactive console. Maybe I can
> > try telnet after netconsole to get a net console 
> > in Linux? Sound a little bit complicated than 
> > what I thought:-)
> 
> So netconsole is a write-only kernel console. 
> Kernel messages will go out, but your init process, 
> etc will not.  For syslog type things, you
> may want to look into a syslogd that does network. 
> I don't know how netconsole support in u-boot 
> works, but it sounds like it's a network
> aware application that you can telnet into, rather
> than Linux's netconsole.

OK. I get it. The netconsole in kernel is different
from the one implemented in u-boot. The u-boot 
netconsole is a interactive console whereas kernel 
netconsole is just a simulated one after booting up.
No wawnder I need to wait 6-8 seconds to get the 
network output after loading kernel.

Thanks so much for your clarification,

Sam 


	

	
		
___________________________________________________________ 
雅虎免费G邮箱－中国第一绝无垃圾邮件骚扰超大邮箱
http://cn.mail.yahoo.com/?id=77071

Re: 2.6.12-rc2 netconsole problem

[Sam Song]

Wolfgang Denk <wd@denx.de> wrote:
[snip]
> It seems you failed to read the
> doc/README.NetConsole fle that  comes
> with  the  U-Boot sources. This explains clearly
> that while in U-Boot the netconsole implementation 
> is bidirectional, the Linux  coede  is
> unidirectional only.

Oops, I read README.NetConsole again and found the 
line in the last part. Sorry for that. I tend to 
make a mistake to do sth in the first time:-)

> > netconsole to get a net console in Linux? Sound a 
> > little bit complicated than what I thought:-)
> 
> You should not use telnet for  obvious  security 
> problems.  Use  ssh instead.  But  from  the 
> technical side it is perfectly fine to use a
> remote login through telnet or rsh or ssh...

OK, good idea to try. 

Thanks very much indeed,

Sam




	

	
		
___________________________________________________________ 
雅虎免费G邮箱－中国第一绝无垃圾邮件骚扰超大邮箱
http://cn.mail.yahoo.com/?id=77071

Re: 2.6.12-rc2 netconsole problem

[Tom Rini]

On Wed, Jul 06, 2005 at 11:12:00AM +0200, Wolfgang Denk wrote:
> In message <20050706051139.GB8358@smtp.west.cox.net> you wrote:
> >
> > may want to look into a syslogd that does network.  I don't know how
> > netconsole support in u-boot works, but it sounds like it's a network
> > aware application that you can telnet into, rather than Linux's
> > netconsole.
> 
> It's an implementation which is compatible with  (and  derived  from)
> Linux'  netconsole,  but  with  an  added  input  channel so you have
> interactive control over ethernet.

That's quite cool.  Any chance that could be ported back to Linux (I
know netpoll supports it, kgdboe works in the kernel).

-- 
Tom Rini
http://gate.crashing.org/~trini/

Re: 2.6.12-rc2 netconsole problem

[Wolfgang Denk]

In message <20050706163029.GC8358@smtp.west.cox.net> you wrote:
> 
> That's quite cool.  Any chance that could be ported back to Linux (I
> know netpoll supports it, kgdboe works in the kernel).

Chances yes, if somebody volunteers or pays for  the  effort.  Sorry,
the old problem: too few hours per day.

Best regards,

Wolfgang Denk

-- 
Software Engineering:  Embedded and Realtime Systems,  Embedded Linux
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd@denx.de
The price one pays for pursuing any profession,  or  calling,  is  an
intimate knowledge of its ugly side.                  - James Baldwin