Date: Sat, 12 Oct 2002 00:22:20 +0200 (MEST) From: Peter Cornelius <pcc@gmx.net> To: freebsd-questions@freebsd.org Cc: freebsd-ports@freebsd.org, Peter Cornelius <pcc@gmx.net> Subject: Anyone running mars_nwe on 4.5-STABLE or later? Message-ID: <20012.1034374940@www7.gmx.net> References: <3643.1034374002@www4.gmx.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi,
is there anyone actually *running* a recent version of mars_nwe that she or
he built from ports? I don't believe it; I had to comment out a few
fprintf()s in order to stop it dump core in the first place in order to make
it start
("wrong" configs still make it dump even though others don't but so far I
did
not have any measurable success).
I've waded through google output for days now, read Boris Popov's small
installation how-to as well as the mars_nwe homepage and mailing list
archives
but there either is something horribly weird with mars or I'm persistently
missing the point.
Below, there's a basic outline of the situation as I see it. Any hints
appreciated, please Cc: me directly in replies. Also, I think the thread
should
stay in questions for the time being (unless there's a better place for it
;-)); I just wanted to catch most of the few possible mars_nwe users so
please
excuse the wide distribution.
Well, thanks a lot for looking at this,
best regards,
Peter.
---
I do have ipx enabled and IPXrouted running. It all seems in order but I
don't know whether I miss some subtle ipx conventions here. From another box
I
do see ipx rip requests but besides that, all is quiet. netstat -fipx -r
says
something I would expect thus I believe this side of the setup is doing what
it should:
Routing tables
IPX:
Destination Gateway Flags Netif Expire
default default U ed1f1
c0a80101.* c0a80101.1 U lo0
As I said, I started nwserv in gdb and found three fprintf()s in tools.c
which I commented out (I'm not an experienced programmer, though). Now
nwserv
starts, reads the config and calls ncpserv which now fails saying it can't
bind:
Mars_nwe V0.99pl19 started using /usr/local/etc/nwserv.conf.
If you have problems, please read mars_nwe/doc/BUGS !
Errors/warnings will be reported in /var/log/nw.log
bash-2.04#
!! 10.11,23:29:17 NCP 0 0:PANIC !!
10.11,23:29:17 NCP 0 0:open_ipx_sockets:Address already in use
start t_bind
!! 10.11,23:29:17 NWS 0 0:PANIC !!
10.11,23:29:17 NWS 0 0:CHILD died:Interrupted system call
Child=NCPSERV, result=1
*********************************************
WARNING: NWE-SERVER shuts down in 10 sec !!!
*********************************************
NWE-SERVER is down now !!
The log file (/var/log/nw.log) now contains the entries:
10.11,23:29:15 NWS 0 0:Starting Version: 0.99pl19
10.11,23:29:15 NWS 0 0:SERVER name='FILESERV', INTERNAL
NET=0xc0a80101, NODE=0x00:00:00:00:00:01
10.11,23:29:15 NWS 0 0:USE_PERMANENT_OUT_SOCKET enabled
10.11,23:29:15 NWS 0 0:IPX_MAX_DATA=1470, RW_BUFFERSIZE =1444
10.11,23:29:15 NWB 0 0:Starting Version: 0.99pl19
10.11,23:29:15 NWB 0 0:check_compress_bindery starts...
10.11,23:29:15 NWB 0 0:check_compress_bindery after qsort
10.11,23:29:15 NWB 0 0:check_compress_bindery after deleting props,
propok=1
10.11,23:29:15 NWB 0 0:check_compress_bindery ends after 0 seconds.
10.11,23:29:15 NWB 0 0:Add/Change User='SUPERVISOR', UnixUser='root'
10.11,23:29:15 NWB 0 0:Add/Change User='GUEST', UnixUser='nobody'
10.11,23:29:15 NWB 0 0:ADD PS(User)=PS0457, Q=PS0457Q
10.11,23:29:15 NWB 0 0:add_pr_server:user=PS0457 not exist
10.11,23:29:15 NWB 0 0:USE_PERMANENT_OUT_SOCKET enabled
10.11,23:29:17 NCP 0 0:Starting Version: 0.99pl19
10.11,23:29:17 NCP 0 0:TLI-BIND:Address already in use
socket Nr:0x451
10.11,23:29:17 NCP 0 0:TLI-BIND:Address already in use
errno: 48
10.11,23:29:17 NCP 0 0:t_error:Address already in use
t_bind,open_ipx_socket
!! 10.11,23:29:17 NCP 0 0:PANIC !!
10.11,23:29:17 NCP 0 0:open_ipx_sockets:Address already in use
!! 10.11,23:29:17 NWS 0 0:PANIC !!
10.11,23:29:17 NWS 0 0:CHILD died:Interrupted system call
Child=NCPSERV, result=1
10.11,23:29:22 NWS 0 0:write_wdata:Interrupted system call
to NCPSERV
All this seems to be independent from the address I set in
${PREFIX}/etc/nwserv.conf (/usr/local/etc/nwserv.conf); I there have (lo0's
ipx address always
adjusted accordingly, if necessary):
# This line (the original one) causes nwserv to dump core:
#3 auto
# These lines produce an equal result:
#3 auto 1
#3 0x7f000001 1
# These lines all cause ncpserv to panic (can't assign address).
#3 0xc0a80101 0x2
#3 0xc0a80101
#3 0x0 1
3 0xC0A80101
What strikes me odd is that when I directly start ncpserv with truss,
sa_data in the actual call seems to be empty in the actual bind call:
bind(0x4,{ sa_len = 0, sa_family = 23, sa_data = { } },16)
ERR#48 'Address already in use'
This seems to be independent of the values I give on the command line.
I have tried to run the port on a 4.5-STABLE box (at home, where all what
you see here comes from) as well as on a 4.7-RC0 (at work, can't do too much
there), with equal insuccess. I also tried the package, but that dumps core
even more than when I build myself.
I'll have another shot at the bugger with a friend who knows how to program
in the next few days while I'll throw in my FreeBSExperinence the little I
have ;-)) However, it may be something utterly simple and obvious, so I
thought
I'd better ask. Any help appreciated as well as hints on what may be worth
while investigating. Ta!
--
+++ GMX - Mail, Messaging & more http://www.gmx.net +++
NEU: Mit GMX ins Internet. Rund um die Uhr für 1 ct/ Min. surfen!
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-ports" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20012.1034374940>