Freetz support for the 5124

olistudent · 14 Apr 2010

Perhaps the port setting is not shown. I think he wrote above that the port is 8081.

MfG Oliver

MaxMuster · 14 Apr 2010

I see, maybe the output is trunkated.
Let's wait for the "netstat" output to see open ports and the processes listening...

karthik · 15 Apr 2010

MaxMuster schrieb:
This one looks strange. If ctlmgr is running (and it should contain the webserver in this firmware) it should use port 80. So how can there be a running httpd w/o a port setting, which should run on port 80, too?!?

So if you could post

Code:

netstat -lpn

Jörg

that was the one I started on 8081. anyways to avoid any confusion I restarted the box, here is the PS and netstat

Code:

/var/mod/root # ps
  PID USER       VSZ STAT COMMAND
    1 root      1144 S    init
    2 root         0 SWN  [ksoftirqd/0]
    3 root         0 SW<  [events/0]
    4 root         0 SW<  [khelper]
    5 root         0 SW<  [kthread]
    6 root         0 SW<  [kblockd/0]
   23 root         0 SW<  [pdflush]
   24 root         0 SW<  [pdflush]
   26 root         0 SW<  [aio/0]
   25 root         0 SW   [kswapd0]
   62 root         0 SW   [pm_info]
   66 root         0 SW<  [CPMAC]
   70 root         0 SW   [mtdblockd]
   96 root         0 SW   [tffsd_mtd_0]
  288 root         0 SW<  [capi_oslib]
  289 root         0 SW<  [capi_oslib]
  290 root         0 SW   [capitransp]
  296 root         0 SW   [glob_codecs]
  309 root         0 SW<  [khubd]
  353 root      6036 S N  ctlmgr
  358 root      2816 S    igdd
  378 root      2520 S    multid
  388 root      6036 S N  ctlmgr
  389 root      6036 S N  ctlmgr
  390 root      6036 S N  ctlmgr
  403 root      2928 S    dsld -i -n
  426 root      1324 S    telnetd -l /sbin/ar7login
  427 root      2084 S    telefon a127.0.0.1
  431 root      4332 S <  voipd
  438 root      2336 S    usermand
  441 root       832 S    /bin/run_clock -c /dev/tffs -d
  478 root      2340 S    /usr/bin/faxd -a
  507 root      1136 S    httpd -P /var/run/webcfg.pid -p 81 -c /mod/etc/httpd
  522 root      1144 S    init
  525 root         0 RWN  [kdsld_token]
  660 root      1156 S    -sh
  696 root      1136 R    ps
/var/mod/root #

Code:

/var/mod/root # netstat -lpn
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address           Foreign Address         State       PID/Program name    
tcp        0      0 0.0.0.0:5060            0.0.0.0:*               LISTEN      431/voipd
tcp        0      0 0.0.0.0:8080            0.0.0.0:*               LISTEN      353/ctlmgr
tcp        0      0 0.0.0.0:80              0.0.0.0:*               LISTEN      353/ctlmgr
tcp        0      0 0.0.0.0:81              0.0.0.0:*               LISTEN      507/httpd
tcp        0      0 127.0.0.1:1011          0.0.0.0:*               LISTEN      427/telefon
tcp        0      0 0.0.0.0:23              0.0.0.0:*               LISTEN      426/telnetd
udp        0      0 0.0.0.0:2048            0.0.0.0:*                           378/multid
udp        0      0 0.0.0.0:2049            0.0.0.0:*                           378/multid
udp        0      0 0.0.0.0:2050            0.0.0.0:*                           353/ctlmgr
udp        0      0 0.0.0.0:2051            0.0.0.0:*                           378/multid
udp        0      0 0.0.0.0:7077            0.0.0.0:*                           431/voipd
udp        0      0 0.0.0.0:53              0.0.0.0:*                           378/multid
udp        0      0 0.0.0.0:5060            0.0.0.0:*                           431/voipd
raw        0      0 0.0.0.0:2               0.0.0.0:*               2           378/multid
raw        0      0 0.0.0.0:2               0.0.0.0:*               2           378/multid
raw        0      0 0.0.0.0:2               0.0.0.0:*               2           378/multid
Active UNIX domain sockets (only servers)
Proto RefCnt Flags       Type       State         I-Node PID/Program name    Path
/var/mod/root #

MaxMuster · 15 Apr 2010

So, this one looks like expected, ctlmgr listens on port 80:

Code:

tcp        0      0 0.0.0.0:80              0.0.0.0:*               LISTEN      353/ctlmgr

And you get "403 forbidden" connecting to the box?

Jörg

karthik · 15 Apr 2010

yes, that is correct, see attached.
thanks

MaxMuster · 15 Apr 2010

Could you please try this two things:

first try should restart AVM-HTTP-server inside ctlmgr in foregorund, so you could see any output maybe generated when you use WebIF (end with <CTRL>C and "ctlmgr" to get it running like before)

second version is use httpd of busybox instead

Code:

# first try
ctlmgr -s; sleep 2; ctlmgr -v -f

# second try
ctlmgr -s; sleep 2; rc.websrv start ; sleep 1 ;ctlmgr

Jörg

EDIT just forget it, take some time (for a "make dirclean" and a new "make") for trying out the new patch in trac (it should resolve a uClibc mismatch causing the error).

karthik · 15 Apr 2010

Still no go, and no additional error messages.

Code:

/var/mod/root # ctlmgr -s ; sleep 2 ; ctlmgr -v -f
Apr 15 21:19:55 faxd[683]: faxd_init, failed to open led-device.
mkdir: can't create directory '/var/samba/private': File exists
1 samba users written to /var/samba/private/smbpasswd

Code:

/var/mod/root # ctlmgr -s ; sleep 2 ; /etc/init.d/rc.websrv start ; sleep 1 ; ct
lmgr
Starting websrv AVM web UI...httpd: chdir(/usr/www/all): No such file or directory
failed.
/var/mod/root # Apr 15 21:24:57 faxd[803]: faxd_init, failed to open led-device.

I was able to start httpd with home pointing to /usr/www/avme and could access web interface from that port.

MaxMuster · 15 Apr 2010

Can you look if there are errors in syslog?
I'm quite sure it is a problem regarding a wrong uclibc version, so if you may try the new patch in the track ticket??
But you will have to recompile a lot, so it will take some time...

Jörg

karthik · 16 Apr 2010

MaxMuster schrieb:
Can you look if there are errors in syslog?
I'm quite sure it is a problem regarding a wrong uclibc version, so if you may try the new patch in the track ticket??
But you will have to recompile a lot, so it will take some time...

Jörg

No, I dont see any error messages in the log.

the new patch bricked the router. or atleast the ethernet doesnt come up.

Code:

$ diff .config .config.old
681,682c681,682
< # FREETZ_TARGET_UCLIBC_VERSION_0_9_28 is not set
< FREETZ_TARGET_UCLIBC_VERSION_0_9_29=y
---
> FREETZ_TARGET_UCLIBC_VERSION_0_9_28=y
> # FREETZ_TARGET_UCLIBC_VERSION_0_9_29 is not set
685,686c685,686
< # FREETZ_TARGET_COMPILER_GCC_4_2_4_UCLIBC_0_9_28 is not set
< FREETZ_TARGET_COMPILER_GCC_4_2_4_UCLIBC_0_9_29=y
---
> FREETZ_TARGET_COMPILER_GCC_4_2_4_UCLIBC_0_9_28=y
> # FREETZ_TARGET_COMPILER_GCC_4_2_4_UCLIBC_0_9_29 is not set
690c690
< FREETZ_TARGET_UCLIBC_VERSION="0.9.29"
---
> FREETZ_TARGET_UCLIBC_VERSION="0.9.28"

olistudent · 16 Apr 2010

Did you do "make dirclean" or start from scratch? Otherwise this could happen atm.

Regards
Oliver

karthik · 16 Apr 2010

olistudent schrieb:
Did you do "make dirclean" or start from scratch? Otherwise this could happen atm.

Regards
Oliver

Yes, I did.

make menuconfig
make dirclean
make

MaxMuster · 16 Apr 2010

We won't get much further here without having the output of the serial console. Do you think you can do this?

Jörg

karthik · 17 Apr 2010

MaxMuster schrieb:
We won't get much further here without having the output of the serial console. Do you think you can do this?

Jörg

hmmm... let me try.

Does avm have any standard for the pin assignment? In the attached figure, I think the square post (#1) is the 3.3v and and #4 is gnd. I dont know which one is Rx, Tx. Would you happen to know?

thanks
Karthik

olistudent · 17 Apr 2010

Code:

   1  2  3  4
  []  o  o  o

   3  R  T  G
   ,  x  x  N
   3  D  D  D
   V

MfG Oliver

MaxMuster · 17 Apr 2010

Yes, I do ;-) Pinout is

Code:

3.3 [] 
Rx  o
Tx  o
GND o

All directions are "from the box-view", so Tx will be the output of the Box, which has to be connected to PCs Input (Rx) and vice versa...

Please note that you can't conntect the Pins directly to your PC's serial but need an converter for the 3.3 V logic of the box. You may use a data cable of a mobile phone or so.

You can even take another Fritzbox as a kind of a serial terminal.

If (as in this case) you only need to see the output of the box, you may try to connect only(!) Tx of the box directly to your PCs serial. If the boxes voltage is sufficient for the PSs com, it might work. Best chances are with a notebook or an usb-to-serial converter. But never connect the PCs Tx to the boxes Rx, it might kill your box.

Joerg

Edit Not the first one, but more detailed ;-)

RalfFriedl · 17 Apr 2010

MaxMuster schrieb:
If you only need to see the output of the box, you may try to connect only(!) Tx of the box directly to your PCs serial.

Have you ever tries this? Not only does the standard specify a voltage of at least +3V/-3V, the converters also act as inverters, that means that 0V->+12V and +3V->-12V. So even if your receiver recognizes 0V as negative Voltage and +3V as positive, the signal is still inverted.

MaxMuster · 17 Apr 2010

No, I havent. I'm quite sure I read about this once, but of course didn't take the inversion into account, so I will take this advice out (and try for myself, to be sure).

Jörg

karthik · 15 Mai 2010

MaxMuster schrieb:
Yes, I do ;-) Pinout is

Code:

3.3 [] Rx o Tx o GND o

All directions are "from the box-view", so Tx will be the output of the Box, which has to be connected to PCs Input (Rx) and vice versa...

Please note that you can't conntect the Pins directly to your PC's serial but need an converter for the 3.3 V logic of the box. You may use a data cable of a mobile phone or so.

You can even take another Fritzbox as a kind of a serial terminal.

If (as in this case) you only need to see the output of the box, you may try to connect only(!) Tx of the box directly to your PCs serial. If the boxes voltage is sufficient for the PSs com, it might work. Best chances are with a notebook or an usb-to-serial converter. But never connect the PCs Tx to the boxes Rx, it might kill your box.

Joerg

Edit Not the first one, but more detailed ;-)

thanks for details.

I am still waiting for the usb-serial converter to arrive. Meanwhile, I was trying to include iptables and selected uClib 0_9_29 and the avm web interface worked this time. may be the "replace kernel" which gets selected with iptables did the trick?

Gompf · 21 Sep 2010

Hi Guys,

(MaxMuster in particular as he is the godfather of 5124 Freetz)

I am writing in English here so others who read this thread might benefit as well. I am using Freetz with the 5124 and it works very well. However recently AVM has released a new Firmware for the 5124 (actually 2+ months ago).

I like to politely ask if anybody who knows how to patch for the 5124 to update the trunk with the new firmware if possible

Thanks !

/Chris

olistudent · 21 Sep 2010

http://trac.freetz.org/changeset/5827

Please test. Especially if replace kernel works would be interesting.

Regards
Oliver

Freetz support for the 5124

IPPF-Urgestein

IPPF-Promi

Neuer User

IPPF-Promi

Neuer User

Anhänge

IPPF-Promi

Neuer User

IPPF-Promi

Neuer User

IPPF-Urgestein

Neuer User

IPPF-Promi

Neuer User

Anhänge

IPPF-Urgestein

IPPF-Promi

IPPF-Urgestein

IPPF-Promi

Neuer User

Neuer User

IPPF-Urgestein

Statistik des Forums