Wireshark · Ethereal-dev: [ethereal-dev] ethereal0.8.1 on HPUX 10.20

[Jost Martin <Martin.Jost@xxxxxxxxxxxxxx>]

Hello

I got ethereal0.8.1 mostly to work on HPUX 10.20.
uname: HP-UX lasagne B.10.20 E 9000/785 2016116051 8-user license

pd; lasagne:/pd-sw/sysadm/ethereal-0.8.1> gtk-config --version
1.2.6

ethereal 0.8.1

First the compilation-hassles:

There are two places, where configure checks for -lpcap. Unfortunately it
first gives -lpcap and then the -L-option to the place where it is. This
will make the test fail.
I changed this:
pd; lasagne:/pd-sw/sysadm/ethereal-0.8.1> diff -c configure.~1~ configure
*** configure.~1~       Sat Jan  8 20:30:44 2000
--- configure   Tue Jan 18 14:07:28 2000
***************
*** 1894,1900 ****
    echo $ac_n "(cached) $ac_c" 1>&6
  else
    ac_save_LIBS="$LIBS"
! LIBS="-lpcap  $LIBS"
  cat > conftest.$ac_ext <<EOF
  #line 1900 "configure"
  #include "confdefs.h"
--- 1894,1900 ----
    echo $ac_n "(cached) $ac_c" 1>&6
  else
    ac_save_LIBS="$LIBS"
! LIBS="$LIBS -lpcap"
  cat > conftest.$ac_ext <<EOF
  #line 1900 "configure"
  #include "confdefs.h"
***************
*** 1928,1934 ****
  #define $ac_tr_lib 1
  EOF
  
!   LIBS="-lpcap $LIBS"
  
  else
    echo "$ac_t""no" 1>&6
--- 1928,1934 ----
  #define $ac_tr_lib 1
  EOF
  
!   LIBS="$LIBS -lpcap"
  
  else
    echo "$ac_t""no" 1>&6


To configure ethereal I need the following CFLAGS:
(I'm using the (optional) ANSI-C Compiler from HP)
export CFLAGS="-Aa -Ae +O2"
-Aa: ANSI C
-Ae: Extensions to ANSI (needed for 'long long')
+O2: Optimization


The resulting Makefile needs to be corrected:

pd; lasagne:/pd-sw/sysadm/ethereal-0.8.1> diff -c Makefile.~1~ Makefile
*** Makefile.~1~        Tue Jan 18 14:07:55 2000
--- Makefile    Tue Jan 18 14:39:14 2000
***************
*** 178,184 ****
  DEFS = -DHAVE_CONFIG_H -I. -I$(srcdir) -I.
  CPPFLAGS =  -I$(top_srcdir) -I$(top_srcdir)/wiretap -I/usr/local/include
  LDFLAGS = 
! LIBS =  -L/usr/local/lib -L/pd/libs/lib -lgtk -lgdk -Wl,-E -lgmodule
-lglib -ldld -lXext -lX11 -lm -lpcap
  ethereal_OBJECTS =  asn1.o capture.o colors.o column.o conversation.o \
  dfilter-grammar.o dfilter-scanner.o dfilter.o ethertype.o file.o \
  follow.o ipproto.o ipv4.o packet.o plugins.o prefs.o print.o proto.o \
--- 178,184 ----
  DEFS = -DHAVE_CONFIG_H -I. -I$(srcdir) -I.
  CPPFLAGS =  -I$(top_srcdir) -I$(top_srcdir)/wiretap -I/usr/local/include
  LDFLAGS = 
! LIBS =  -L/usr/local/lib -L/pd/libs/lib -lgtk -lgdk -Wl,-E -lgmodule
-lglib -ldld -lXext -lX11 -lm -lpcap -lc -lPW
  ethereal_OBJECTS =  asn1.o capture.o colors.o column.o conversation.o \
  dfilter-grammar.o dfilter-scanner.o dfilter.o ethertype.o file.o \
  follow.o ipproto.o ipv4.o packet.o plugins.o prefs.o print.o proto.o \
***************
*** 961,967 ****
        ./rdps $(srcdir)/print.ps ps.c
  
  rdps: rdps.c
!       $(CC) -o rdps $(srcdir)/rdps.c
  
  editcap.o: editcap.c
        $(COMPILE) -c $(srcdir)/editcap.c
--- 961,967 ----
        ./rdps $(srcdir)/print.ps ps.c
  
  rdps: rdps.c
!       $(CC) $(CFLAGS) -o rdps $(srcdir)/rdps.c
  
  editcap.o: editcap.c
        $(COMPILE) -c $(srcdir)/editcap.c

On LIBS '-lc -lPW' needs to be appended. -lPW supplies alloca(). -lc before
-lPW is needed to work around a bug in PW (at least on HPUX 10.20), else
Error(data) remains undef.
The rule for rdps is missing the CFLAGS (thus failing to compile)

I don't have a header dlfcn.h (what is this suposed to be ?)
If this header isn't found, the contents of plugins.c is defined away,
resulting in init_plugins() being undef.
I therefore commented the call to init_plugins() in main.c.
(I would like to know what's going on here - this seems to be clumsy...)

Usage-problem:
I want to use nettl to gather network-Info. I gather the data with:
nettl -tn pduin pduout -e ns_ls_nfs ns_ls_udp ns_ls_tcp -file ~/TRC
The I stop the tracing with
nettl -traceoff -e all  

When I load the resulting trace-File in ethereal (File/load) nothing
happens, except that the filename appears in the lower right corner with
"Drops: 0".
The trace-file itself seems to be ok. If I format the file using HPs netfmt
I get output:
[lasagne_ROOT]:>netfmt -N -f TRC.TRC0
 
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ARPA/9000
NETWORKING^^^^^^^^^^^^^^^^^^^^^^^^^^@#%
  Timestamp            : Tue Jan 18 MET 2000 15:26:54.693634
  Process ID           : [ICS]              Subsystem        : NS_LS_UDP
  User ID ( UID )      : -1                 Trace Kind       : PDU IN TRACE
  Device ID            : -1                 Path ID          : -1
  Connection ID        : 0
  Location             : 00123
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-------------------------------- UDP Header
----------------------------------
sport:   1008   -->   dport:  nfsd     data len: 120   chksum: 0xbb49 
-------------------------------- RPC Call
------------------------------------
trans id: 0x387ab587 rpc version: 2    prog: 100003   version: 3    proc: 6
auth type: UNIX      machine: garfield                uid: 208      gid: 20
verf type: 0         verf length: 0 
-------------------------------- NFS3 Call
-----------------------------------
proc     : READ        file handle type: HP  
file handle:
   0: 40 02 00 01 ff ff ff ff 00 0a 00 00 00 00 5c 7b  @.............\{
  16: 00 00 00 08 00 0a 00 00 00 00 37 1f 00 00 00 01  ..........7.....
offset: 6840320 count: 8192         
 
vvvvvvvvvvvvvvvvvvvvvvvvvvvvvvARPA/9000
NETWORKINGvvvvvvvvvvvvvvvvvvvvvvvvvv@#%
  Timestamp            : Tue Jan 18 MET 2000 15:26:54.693773
  Process ID           : 1290               Subsystem        : NS_LS_UDP
  User ID ( UID )      : 0                  Trace Kind       : PDU OUT TRACE
  Device ID            : -1                 Path ID          : -1
  Connection ID        : 0
  Location             : 00123
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-------------------------------- UDP Header
----------------------------------
sport:   nfsd   -->   dport:  1008     data len: 8320  chksum: 0x763e 
-------------------------------- RPC Reply
-----------------------------------
trans id: 0x387ab587       Accepted      verif type: 0      verif length: 0
-------------------------------- NFS3 Reply
----------------------------------
proc     : READ        status : OK
file type: 1           mode   : 00740            hard links: 1       
uid      : 208         gid    : 20               size      : 7162179
used     : 57303040    major #: 0                minor #   : 0
inode #  : 23675       fsid   : 0x40020001
accessed : Tue Jan 18 15:26:54.685751  modified: Mon Nov 22 14:57:59.000000
changed  : Sun Jan 16 08:15:59.853847
count    : 8192        EOF : FALSE 
-------------------------------- Read Data
-----------------------------------
   0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff  ................
 < printing suppressed for one or more repetitions of the previous line >

...

I need nettl for some of our machines, because they use STREAMS not DLPI...
(And I much prefer ethereal-display to the netfmt-output shown above)

Martin Jost

P.S.:
Thank you for ethereal - real great to figure out whats going on !
(Or just to learn about networking)