Received: from relay2.fe.up.pt (relay2.fe.up.pt [193.136.28.34]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l7TNsDC03333 for ; Wed, 29 Aug 2007 16:54:14 -0700 Received: from eva1.fe.up.pt (EVA1.fe.up.pt [192.168.50.157]) by relay2.fe.up.pt (8.13.4/8.13.1) with ESMTP id l7U0rlld012076 for ; Thu, 30 Aug 2007 01:53:58 +0100 Received: from me262.hell (tcpgate.fe.up.pt [192.168.50.18]) by eva1.fe.up.pt (Postfix) with ESMTP id 282F118D3 for ; Thu, 30 Aug 2007 01:53:47 +0100 (WEST) Message-ID: <46D6154A.1000209@fe.up.pt> Date: Thu, 30 Aug 2007 01:54:34 +0100 From: Ricardo Martins User-Agent: Thunderbird 2.0.0.5 (X11/20070719) MIME-Version: 1.0 To: dtn-users@mailman.dtnrg.org Subject: Re: [dtn-users] dtn on arm References: <52950c3f0708141326v4f7ec5d6ie6588ed222ab7d0@mail.gmail.com> <52950c3f0708150721v561c3071t16ce6874cb73fe2d@mail.gmail.com> <890FE8E0-AD2A-4767-829A-FC3669CCC37B@cs.berkeley.edu> In-Reply-To: <890FE8E0-AD2A-4767-829A-FC3669CCC37B@cs.berkeley.edu> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-Virus-Scanned: by amavisd-new Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: Hi, This website explains the cause of this problem: http://www.arm.linux.org.uk/docs/faqs/signedchar.php Regards, Ricardo Martins Michael Demmer wrote: > No need for a patch -- I committed the fixes. Let me know if I missed > anything. > > -m > > On Aug 15, 2007, at 7:21 AM, Timur Alperovich wrote: > >> Right, that works :) >> That was the first solution I was thinking of. Do you want patches for >> that? It's a minor fix really. >> >> Timur >> >> On 8/14/07, *Michael Demmer * > > wrote: >> >> >> Can you try changing the "char c" declaration into "int c" and see >> if that fixes the problem? getopt is supposed to return an int of >> -1 when reaching the end of the option string, so my guess is that >> the incompatibility occurs when casting that int to a char. >> >> -m >> >> On Aug 14, 2007, at 1:26 PM, Timur Alperovich wrote: >> >>> Hi, >>> >>> I've been putting dtn on small nodes, gumstix (www.gumstix.com >>> ). What I noticed is that I could not run >>> the dtnsend, dtnrecv commands as well as a few others. The >>> commands would simply print out the usage information regardless >>> of the arguments and no errors. It turned out in the code for >>> these there is an assignment such as "char c = getopt(...);", >>> which results in c = 255 when getopt returns -1 on completion on >>> the arm7. I found there are two ways to avoid this: 1. use code >>> similar to that dtnperf: int c; while((c = getopt(...)) != EOF); >>> 2. add case 255 in the switch statement that also results in the >>> "done" condition. I think the first approach is preferred. >>> Anyway, I wanted to point this out for anyone who is doing >>> anything similar. I can submit patches for files to change the >>> parsing if that would be useful. >>> >>> Timur >> >> > Received: from n066.sc1.he.tucows.com (smtpout0150.sc1.he.tucows.com [64.97.136.150]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l7JMbRC04171 for ; Sun, 19 Aug 2007 15:37:27 -0700 Received: from [192.168.0.6] (86.31.126.90) by n066.sc1.he.tucows.com (7.2.069.1) id 466CB207004DBFD6 for dtn-users@mailman.dtnrg.org; Sun, 19 Aug 2007 23:37:20 +0000 Mime-Version: 1.0 (Apple Message framework v752.2) In-Reply-To: References: <9CE6073D-FDD4-404E-BCF9-A0967BA754FE@virgin.net> Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: Content-Transfer-Encoding: 7bit From: Darren Long Subject: Re: [dtn-users] DTN2 configure problems with BerkeleyDB using distcc Date: Mon, 20 Aug 2007 00:37:15 +0100 To: dtn-users@mailman.dtnrg.org X-Mailer: Apple Mail (2.752.2) Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: Hi Mike, Thanks very much. configure succeeds now with either method, i.e. the exports or the switches to force use of distcc. I think Macs support distcc. I've only got one, so there's no point me trying, but if you have a few, it may be worth a go. Cheers, Darren On 19 Aug 2007, at 23:05, Michael Demmer wrote: > Ah sure.... it's because the configure script tries to infer the C+ > + compiler to use based on the C compiler settings: > > ac_try_cxx=`echo $ac_with_cc | sed 's/cc/++/'` > > I just checked in a change to allow explicit setting of the C++ > compiler based on the CXX environment variable or the --with-cxx > configure option, with a fallback to use the same inferred setting > as before if the cxx compiler isn't supplied. > > Update and let me know if that works better for you. > > -m > > > On Aug 19, 2007, at 1:12 PM, Darren Long wrote: > >> On 19 Aug 2007, at 19:41, Michael Demmer wrote: >>> --with-cc="distcc arm-linux-gnu-gcc" >> >> >> Hi Mike, >> >> I had a go with that, no joy, but got some clues. I went back to >> using the exports, and configure does pick up the CC correctly, >> but seems to ignore CXX and tries to make something up for the C++ >> compiler name - "dist++", at configure:4060. This seems to be the >> reason for the configure problem with BerkeleyDB, as the tests use >> the c++ compiler, and fail on the missing binary. >> >> >> Some output from configure.log is below, called with ./configure -- >> disable-ecl --disable-edp --with-db=/usr/local/BerkeleyDB.4.5 >> >> Cheers, >> >> Darren >> >> ## ----------- ## >> ## Core tests. ## >> ## ----------- ## >> >> configure:1860: checking location of source directory >> configure:1863: result: "/home/darren/source/DTN2" >> configure:1868: checking location of build directory >> configure:1870: result: "/home/darren/source/DTN2" >> configure:1972: checking for gcc >> configure:1999: result: distcc arm-linux-gnu-gcc >> configure:2237: checking for C compiler version >> configure:2244: distcc arm-linux-gnu-gcc --version >&5 >> arm-linux-gnu-gcc (GCC) 4.1.2 20061115 (prerelease) (Debian 4.1.1-21) >> Copyright (C) 2006 Free Software Foundation, Inc. >> This is free software; see the source for copying conditions. >> There is NO >> warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR >> PURPOSE. >> >> configure:2247: $? = 0 >> configure:2254: distcc arm-linux-gnu-gcc -v >&5 >> Using built-in specs. >> Target: arm-linux-gnu >> Configured with: ../src/configure -v --enable-languages=c,c+ >> +,fortran,objc,obj-c++,treelang --prefix=/usr --enable-shared -- >> with-system-zlib --libexecdir=/usr/lib --without-included-gettext >> --enable-threads=posix --enable-nls --program-suffix=-4.1 --enable- >> __cxa_atexit --enable-clocale=gnu --enable-libstdcxx-debug -- >> enable-mpfr --enable-checking=release arm-linux-gnu >> Thread model: posix >> gcc version 4.1.2 20061115 (prerelease) (Debian 4.1.1-21) >> configure:2257: $? = 0 >> configure:2264: distcc arm-linux-gnu-gcc -V >&5 >> arm-linux-gnu-gcc: '-V' option must have argument >> distcc[26074] ERROR: compile (null) on localhost failed >> configure:2267: $? = 1 >> configure:2290: checking for C compiler default output file name >> configure:2317: distcc arm-linux-gnu-gcc conftest.c >&5 >> configure:2320: $? = 0 >> configure:2358: result: a.out >> configure:2375: checking whether the C compiler works >> configure:2385: ./a.out >> configure:2388: $? = 0 >> configure:2405: result: yes >> configure:2412: checking whether we are cross compiling >> configure:2414: result: no >> configure:2417: checking for suffix of executables >> configure:2424: distcc arm-linux-gnu-gcc -o conftest >> conftest.c >&5 >> configure:2427: $? = 0 >> configure:2451: result: >> configure:2457: checking for suffix of object files >> configure:2483: distcc arm-linux-gnu-gcc -c conftest.c >&5 >> configure:2486: $? = 0 >> configure:2509: result: o >> configure:2513: checking whether we are using the GNU C compiler >> configure:2542: distcc arm-linux-gnu-gcc -c conftest.c >&5 >> configure:2548: $? = 0 >> configure:2565: result: yes >> configure:2570: checking whether distcc arm-linux-gnu-gcc accepts -g >> configure:2600: distcc arm-linux-gnu-gcc -c -g conftest.c >&5 >> configure:2606: $? = 0 >> configure:2705: result: yes >> configure:2722: checking for distcc arm-linux-gnu-gcc option to >> accept ISO C89 >> configure:2796: distcc arm-linux-gnu-gcc -c -g -O2 conftest.c >&5 >> configure:2802: $? = 0 >> configure:2825: result: none needed >> configure:2907: checking for ar >> configure:2923: found /usr/bin/ar >> configure:2934: result: ar >> configure:3003: checking for ranlib >> configure:3019: found /usr/bin/ranlib >> configure:3030: result: ranlib >> configure:3071: checking whether to compile with debugging >> configure:3073: result: yes >> configure:3092: checking whether to compile with memory debugging >> configure:3094: result: no >> configure:3113: checking whether to compile with lock debugging >> (default yes) >> configure:3115: result: yes >> configure:3135: checking whether to compile with optimization >> configure:3137: result: no >> configure:3158: checking whether to compile with profiling >> configure:3160: result: no >> configure:3178: checking whether to link statically >> configure:3180: result: no >> configure:3191: checking whether to link using static external >> libraries >> configure:3193: result: no >> configure:3205: checking whether to try to build shared libraries >> configure:3220: result: yes >> configure:3231: checking whether the compiler supports -fPIC >> configure:3242: distcc arm-linux-gnu-gcc -c -fPIC -DPIC -g -O2 - >> Werror conftest.c >&5 >> configure:3248: $? = 0 >> configure:3262: result: yes >> configure:3268: checking whether the compiler can link a dynamic >> library with -shared >> configure:3280: distcc arm-linux-gnu-gcc -o conftest -fPIC -DPIC - >> g -O2 -Werror -shared conftest.c >&5 >> configure:3286: $? = 0 >> configure:3302: result: yes >> configure:3310: checking extension for dynamic libraries >> configure:3322: result: so >> configure:3325: checking if the compiler supports -Wl,--rpath=. >> configure:3337: distcc arm-linux-gnu-gcc -o conftest -fPIC -DPIC - >> g -O2 -Werror -shared -Wl,-rpath=/foo conftest.c >&5 >> configure:3343: $? = 0 >> configure:3359: result: yes >> configure:3413: checking for grep that handles long lines and -e >> configure:3487: result: /bin/grep >> configure:3492: checking for egrep >> configure:3570: result: /bin/grep -E >> configure:3576: checking for a C compiler (trying distcc arm-linux- >> gnu-gcc) >> configure:3578: result: >> configure:3697: checking for C compiler version >> configure:3704: distcc arm-linux-gnu-gcc --version >&5 >> arm-linux-gnu-gcc (GCC) 4.1.2 20061115 (prerelease) (Debian 4.1.1-21) >> Copyright (C) 2006 Free Software Foundation, Inc. >> This is free software; see the source for copying conditions. >> There is NO >> warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR >> PURPOSE. >> >> configure:3707: $? = 0 >> configure:3714: distcc arm-linux-gnu-gcc -v >&5 >> Using built-in specs. >> Target: arm-linux-gnu >> Configured with: ../src/configure -v --enable-languages=c,c+ >> +,fortran,objc,obj-c++,treelang --prefix=/usr --enable-shared -- >> with-system-zlib --libexecdir=/usr/lib --without-included-gettext >> --enable-threads=posix --enable-nls --program-suffix=-4.1 --enable- >> __cxa_atexit --enable-clocale=gnu --enable-libstdcxx-debug -- >> enable-mpfr --enable-checking=release arm-linux-gnu >> Thread model: posix >> gcc version 4.1.2 20061115 (prerelease) (Debian 4.1.1-21) >> configure:3717: $? = 0 >> configure:3724: distcc arm-linux-gnu-gcc -V >&5 >> arm-linux-gnu-gcc: '-V' option must have argument >> distcc[26235] ERROR: compile (null) on localhost failed >> configure:3727: $? = 1 >> configure:3730: checking whether we are using the GNU C compiler >> configure:3782: result: yes >> configure:3787: checking whether distcc arm-linux-gnu-gcc accepts -g >> configure:3922: result: yes >> configure:3939: checking for distcc arm-linux-gnu-gcc option to >> accept ISO C89 >> configure:4042: result: none needed >> configure:4060: checking for a C++ compiler (trying dist++ arm- >> linux-gnu-gcc) >> configure:4062: result: >> configure:4180: checking for C++ compiler version >> configure:4187: dist++ arm-linux-gnu-gcc --version >&5 >> ./configure: line 4188: dist++: command not found >> configure:4190: $? = 127 >> configure:4197: dist++ arm-linux-gnu-gcc -v >&5 >> ./configure: line 4198: dist++: command not found >> configure:4200: $? = 127 >> configure:4207: dist++ arm-linux-gnu-gcc -V >&5 >> ./configure: line 4208: dist++: command not found >> configure:4210: $? = 127 >> configure:4213: checking whether we are using the GNU C++ compiler >> configure:4242: dist++ arm-linux-gnu-gcc -c conftest.cpp >&5 >> ./configure: line 4243: dist++: command not found >> configure:4248: $? = 127 >> configure: failed program was: >> | /* confdefs.h. */ >> | #define PACKAGE_NAME "" >> | #define PACKAGE_TARNAME "" >> | #define PACKAGE_VERSION "" >> | #define PACKAGE_STRING "" >> | #define PACKAGE_BUGREPORT "" >> | #define HAVE_CONFIG_STATE 1 >> | #define INSTALL_SYSCONFDIR "/etc" >> | #define INSTALL_LOCALSTATEDIR "/var" >> | #define OASYS_DEBUG_LOCKING_ENABLED 1 >> | /* end confdefs.h. */ >> | >> | int >> | main () >> | { >> | #ifndef __GNUC__ >> | choke me >> | #endif >> | >> | ; >> | return 0; >> | } >> configure:4265: result: no >> configure:4270: checking whether dist++ arm-linux-gnu-gcc accepts -g >> configure:4300: dist++ arm-linux-gnu-gcc -c -g conftest.cpp >&5 >> ./configure: line 4301: dist++: command not found >> configure:4306: $? = 127 >> configure: failed program was: >> | /* confdefs.h. */ >> >> >> >> _______________________________________________ >> dtn-users mailing list >> dtn-users@mailman.dtnrg.org >> http://mailman.dtnrg.org/mailman/listinfo/dtn-users > > _______________________________________________ > dtn-users mailing list > dtn-users@mailman.dtnrg.org > http://mailman.dtnrg.org/mailman/listinfo/dtn-users Received: from gateway0.EECS.Berkeley.EDU (gateway0.EECS.Berkeley.EDU [169.229.60.93]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l7JL5sC03583 for ; Sun, 19 Aug 2007 14:05:54 -0700 Received: from [192.168.1.101] (c-67-170-220-149.hsd1.ca.comcast.net [67.170.220.149]) (authenticated bits=0) by gateway0.EECS.Berkeley.EDU (8.14.1/8.13.5) with ESMTP id l7JM5rsO015398 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Sun, 19 Aug 2007 15:05:54 -0700 (PDT) In-Reply-To: References: <9CE6073D-FDD4-404E-BCF9-A0967BA754FE@virgin.net> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: Content-Transfer-Encoding: 7bit From: Michael Demmer Subject: Re: [dtn-users] DTN2 configure problems with BerkeleyDB using distcc X-Applemailsentby: demmer Date: Sun, 19 Aug 2007 15:05:58 -0700 To: dtn-users@mailman.dtnrg.org X-Mailer: Apple Mail (2.752.3) Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: Ah sure.... it's because the configure script tries to infer the C++ compiler to use based on the C compiler settings: ac_try_cxx=`echo $ac_with_cc | sed 's/cc/++/'` I just checked in a change to allow explicit setting of the C++ compiler based on the CXX environment variable or the --with-cxx configure option, with a fallback to use the same inferred setting as before if the cxx compiler isn't supplied. Update and let me know if that works better for you. -m On Aug 19, 2007, at 1:12 PM, Darren Long wrote: > On 19 Aug 2007, at 19:41, Michael Demmer wrote: >> --with-cc="distcc arm-linux-gnu-gcc" > > > Hi Mike, > > I had a go with that, no joy, but got some clues. I went back to > using the exports, and configure does pick up the CC correctly, but > seems to ignore CXX and tries to make something up for the C++ > compiler name - "dist++", at configure:4060. This seems to be the > reason for the configure problem with BerkeleyDB, as the tests use > the c++ compiler, and fail on the missing binary. > > > Some output from configure.log is below, called with ./configure -- > disable-ecl --disable-edp --with-db=/usr/local/BerkeleyDB.4.5 > > Cheers, > > Darren > > ## ----------- ## > ## Core tests. ## > ## ----------- ## > > configure:1860: checking location of source directory > configure:1863: result: "/home/darren/source/DTN2" > configure:1868: checking location of build directory > configure:1870: result: "/home/darren/source/DTN2" > configure:1972: checking for gcc > configure:1999: result: distcc arm-linux-gnu-gcc > configure:2237: checking for C compiler version > configure:2244: distcc arm-linux-gnu-gcc --version >&5 > arm-linux-gnu-gcc (GCC) 4.1.2 20061115 (prerelease) (Debian 4.1.1-21) > Copyright (C) 2006 Free Software Foundation, Inc. > This is free software; see the source for copying conditions. > There is NO > warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR > PURPOSE. > > configure:2247: $? = 0 > configure:2254: distcc arm-linux-gnu-gcc -v >&5 > Using built-in specs. > Target: arm-linux-gnu > Configured with: ../src/configure -v --enable-languages=c,c+ > +,fortran,objc,obj-c++,treelang --prefix=/usr --enable-shared -- > with-system-zlib --libexecdir=/usr/lib --without-included-gettext -- > enable-threads=posix --enable-nls --program-suffix=-4.1 --enable- > __cxa_atexit --enable-clocale=gnu --enable-libstdcxx-debug --enable- > mpfr --enable-checking=release arm-linux-gnu > Thread model: posix > gcc version 4.1.2 20061115 (prerelease) (Debian 4.1.1-21) > configure:2257: $? = 0 > configure:2264: distcc arm-linux-gnu-gcc -V >&5 > arm-linux-gnu-gcc: '-V' option must have argument > distcc[26074] ERROR: compile (null) on localhost failed > configure:2267: $? = 1 > configure:2290: checking for C compiler default output file name > configure:2317: distcc arm-linux-gnu-gcc conftest.c >&5 > configure:2320: $? = 0 > configure:2358: result: a.out > configure:2375: checking whether the C compiler works > configure:2385: ./a.out > configure:2388: $? = 0 > configure:2405: result: yes > configure:2412: checking whether we are cross compiling > configure:2414: result: no > configure:2417: checking for suffix of executables > configure:2424: distcc arm-linux-gnu-gcc -o conftest conftest.c > >&5 > configure:2427: $? = 0 > configure:2451: result: > configure:2457: checking for suffix of object files > configure:2483: distcc arm-linux-gnu-gcc -c conftest.c >&5 > configure:2486: $? = 0 > configure:2509: result: o > configure:2513: checking whether we are using the GNU C compiler > configure:2542: distcc arm-linux-gnu-gcc -c conftest.c >&5 > configure:2548: $? = 0 > configure:2565: result: yes > configure:2570: checking whether distcc arm-linux-gnu-gcc accepts -g > configure:2600: distcc arm-linux-gnu-gcc -c -g conftest.c >&5 > configure:2606: $? = 0 > configure:2705: result: yes > configure:2722: checking for distcc arm-linux-gnu-gcc option to > accept ISO C89 > configure:2796: distcc arm-linux-gnu-gcc -c -g -O2 conftest.c >&5 > configure:2802: $? = 0 > configure:2825: result: none needed > configure:2907: checking for ar > configure:2923: found /usr/bin/ar > configure:2934: result: ar > configure:3003: checking for ranlib > configure:3019: found /usr/bin/ranlib > configure:3030: result: ranlib > configure:3071: checking whether to compile with debugging > configure:3073: result: yes > configure:3092: checking whether to compile with memory debugging > configure:3094: result: no > configure:3113: checking whether to compile with lock debugging > (default yes) > configure:3115: result: yes > configure:3135: checking whether to compile with optimization > configure:3137: result: no > configure:3158: checking whether to compile with profiling > configure:3160: result: no > configure:3178: checking whether to link statically > configure:3180: result: no > configure:3191: checking whether to link using static external > libraries > configure:3193: result: no > configure:3205: checking whether to try to build shared libraries > configure:3220: result: yes > configure:3231: checking whether the compiler supports -fPIC > configure:3242: distcc arm-linux-gnu-gcc -c -fPIC -DPIC -g -O2 - > Werror conftest.c >&5 > configure:3248: $? = 0 > configure:3262: result: yes > configure:3268: checking whether the compiler can link a dynamic > library with -shared > configure:3280: distcc arm-linux-gnu-gcc -o conftest -fPIC -DPIC -g > -O2 -Werror -shared conftest.c >&5 > configure:3286: $? = 0 > configure:3302: result: yes > configure:3310: checking extension for dynamic libraries > configure:3322: result: so > configure:3325: checking if the compiler supports -Wl,--rpath=. > configure:3337: distcc arm-linux-gnu-gcc -o conftest -fPIC -DPIC -g > -O2 -Werror -shared -Wl,-rpath=/foo conftest.c >&5 > configure:3343: $? = 0 > configure:3359: result: yes > configure:3413: checking for grep that handles long lines and -e > configure:3487: result: /bin/grep > configure:3492: checking for egrep > configure:3570: result: /bin/grep -E > configure:3576: checking for a C compiler (trying distcc arm-linux- > gnu-gcc) > configure:3578: result: > configure:3697: checking for C compiler version > configure:3704: distcc arm-linux-gnu-gcc --version >&5 > arm-linux-gnu-gcc (GCC) 4.1.2 20061115 (prerelease) (Debian 4.1.1-21) > Copyright (C) 2006 Free Software Foundation, Inc. > This is free software; see the source for copying conditions. > There is NO > warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR > PURPOSE. > > configure:3707: $? = 0 > configure:3714: distcc arm-linux-gnu-gcc -v >&5 > Using built-in specs. > Target: arm-linux-gnu > Configured with: ../src/configure -v --enable-languages=c,c+ > +,fortran,objc,obj-c++,treelang --prefix=/usr --enable-shared -- > with-system-zlib --libexecdir=/usr/lib --without-included-gettext -- > enable-threads=posix --enable-nls --program-suffix=-4.1 --enable- > __cxa_atexit --enable-clocale=gnu --enable-libstdcxx-debug --enable- > mpfr --enable-checking=release arm-linux-gnu > Thread model: posix > gcc version 4.1.2 20061115 (prerelease) (Debian 4.1.1-21) > configure:3717: $? = 0 > configure:3724: distcc arm-linux-gnu-gcc -V >&5 > arm-linux-gnu-gcc: '-V' option must have argument > distcc[26235] ERROR: compile (null) on localhost failed > configure:3727: $? = 1 > configure:3730: checking whether we are using the GNU C compiler > configure:3782: result: yes > configure:3787: checking whether distcc arm-linux-gnu-gcc accepts -g > configure:3922: result: yes > configure:3939: checking for distcc arm-linux-gnu-gcc option to > accept ISO C89 > configure:4042: result: none needed > configure:4060: checking for a C++ compiler (trying dist++ arm- > linux-gnu-gcc) > configure:4062: result: > configure:4180: checking for C++ compiler version > configure:4187: dist++ arm-linux-gnu-gcc --version >&5 > ./configure: line 4188: dist++: command not found > configure:4190: $? = 127 > configure:4197: dist++ arm-linux-gnu-gcc -v >&5 > ./configure: line 4198: dist++: command not found > configure:4200: $? = 127 > configure:4207: dist++ arm-linux-gnu-gcc -V >&5 > ./configure: line 4208: dist++: command not found > configure:4210: $? = 127 > configure:4213: checking whether we are using the GNU C++ compiler > configure:4242: dist++ arm-linux-gnu-gcc -c conftest.cpp >&5 > ./configure: line 4243: dist++: command not found > configure:4248: $? = 127 > configure: failed program was: > | /* confdefs.h. */ > | #define PACKAGE_NAME "" > | #define PACKAGE_TARNAME "" > | #define PACKAGE_VERSION "" > | #define PACKAGE_STRING "" > | #define PACKAGE_BUGREPORT "" > | #define HAVE_CONFIG_STATE 1 > | #define INSTALL_SYSCONFDIR "/etc" > | #define INSTALL_LOCALSTATEDIR "/var" > | #define OASYS_DEBUG_LOCKING_ENABLED 1 > | /* end confdefs.h. */ > | > | int > | main () > | { > | #ifndef __GNUC__ > | choke me > | #endif > | > | ; > | return 0; > | } > configure:4265: result: no > configure:4270: checking whether dist++ arm-linux-gnu-gcc accepts -g > configure:4300: dist++ arm-linux-gnu-gcc -c -g conftest.cpp >&5 > ./configure: line 4301: dist++: command not found > configure:4306: $? = 127 > configure: failed program was: > | /* confdefs.h. */ > > > > _______________________________________________ > dtn-users mailing list > dtn-users@mailman.dtnrg.org > http://mailman.dtnrg.org/mailman/listinfo/dtn-users Received: from n064.sc1.he.tucows.com (smtpout0160.sc1.he.tucows.com [64.97.136.160]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l7JJCwC02893 for ; Sun, 19 Aug 2007 12:12:58 -0700 Received: from [192.168.0.6] (86.31.126.90) by n064.sc1.he.tucows.com (7.2.069.1) id 466CB1C800523F60 for dtn-users@mailman.dtnrg.org; Sun, 19 Aug 2007 20:12:52 +0000 Mime-Version: 1.0 (Apple Message framework v752.2) In-Reply-To: References: <9CE6073D-FDD4-404E-BCF9-A0967BA754FE@virgin.net> Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: Content-Transfer-Encoding: 7bit From: Darren Long Subject: Re: [dtn-users] DTN2 configure problems with BerkeleyDB using distcc Date: Sun, 19 Aug 2007 21:12:49 +0100 To: dtn-users@mailman.dtnrg.org X-Mailer: Apple Mail (2.752.2) Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: On 19 Aug 2007, at 19:41, Michael Demmer wrote: > --with-cc="distcc arm-linux-gnu-gcc" Hi Mike, I had a go with that, no joy, but got some clues. I went back to using the exports, and configure does pick up the CC correctly, but seems to ignore CXX and tries to make something up for the C++ compiler name - "dist++", at configure:4060. This seems to be the reason for the configure problem with BerkeleyDB, as the tests use the c++ compiler, and fail on the missing binary. Some output from configure.log is below, called with ./configure -- disable-ecl --disable-edp --with-db=/usr/local/BerkeleyDB.4.5 Cheers, Darren ## ----------- ## ## Core tests. ## ## ----------- ## configure:1860: checking location of source directory configure:1863: result: "/home/darren/source/DTN2" configure:1868: checking location of build directory configure:1870: result: "/home/darren/source/DTN2" configure:1972: checking for gcc configure:1999: result: distcc arm-linux-gnu-gcc configure:2237: checking for C compiler version configure:2244: distcc arm-linux-gnu-gcc --version >&5 arm-linux-gnu-gcc (GCC) 4.1.2 20061115 (prerelease) (Debian 4.1.1-21) Copyright (C) 2006 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. configure:2247: $? = 0 configure:2254: distcc arm-linux-gnu-gcc -v >&5 Using built-in specs. Target: arm-linux-gnu Configured with: ../src/configure -v --enable-languages=c,c+ +,fortran,objc,obj-c++,treelang --prefix=/usr --enable-shared --with- system-zlib --libexecdir=/usr/lib --without-included-gettext --enable- threads=posix --enable-nls --program-suffix=-4.1 --enable- __cxa_atexit --enable-clocale=gnu --enable-libstdcxx-debug --enable- mpfr --enable-checking=release arm-linux-gnu Thread model: posix gcc version 4.1.2 20061115 (prerelease) (Debian 4.1.1-21) configure:2257: $? = 0 configure:2264: distcc arm-linux-gnu-gcc -V >&5 arm-linux-gnu-gcc: '-V' option must have argument distcc[26074] ERROR: compile (null) on localhost failed configure:2267: $? = 1 configure:2290: checking for C compiler default output file name configure:2317: distcc arm-linux-gnu-gcc conftest.c >&5 configure:2320: $? = 0 configure:2358: result: a.out configure:2375: checking whether the C compiler works configure:2385: ./a.out configure:2388: $? = 0 configure:2405: result: yes configure:2412: checking whether we are cross compiling configure:2414: result: no configure:2417: checking for suffix of executables configure:2424: distcc arm-linux-gnu-gcc -o conftest conftest.c >&5 configure:2427: $? = 0 configure:2451: result: configure:2457: checking for suffix of object files configure:2483: distcc arm-linux-gnu-gcc -c conftest.c >&5 configure:2486: $? = 0 configure:2509: result: o configure:2513: checking whether we are using the GNU C compiler configure:2542: distcc arm-linux-gnu-gcc -c conftest.c >&5 configure:2548: $? = 0 configure:2565: result: yes configure:2570: checking whether distcc arm-linux-gnu-gcc accepts -g configure:2600: distcc arm-linux-gnu-gcc -c -g conftest.c >&5 configure:2606: $? = 0 configure:2705: result: yes configure:2722: checking for distcc arm-linux-gnu-gcc option to accept ISO C89 configure:2796: distcc arm-linux-gnu-gcc -c -g -O2 conftest.c >&5 configure:2802: $? = 0 configure:2825: result: none needed configure:2907: checking for ar configure:2923: found /usr/bin/ar configure:2934: result: ar configure:3003: checking for ranlib configure:3019: found /usr/bin/ranlib configure:3030: result: ranlib configure:3071: checking whether to compile with debugging configure:3073: result: yes configure:3092: checking whether to compile with memory debugging configure:3094: result: no configure:3113: checking whether to compile with lock debugging (default yes) configure:3115: result: yes configure:3135: checking whether to compile with optimization configure:3137: result: no configure:3158: checking whether to compile with profiling configure:3160: result: no configure:3178: checking whether to link statically configure:3180: result: no configure:3191: checking whether to link using static external libraries configure:3193: result: no configure:3205: checking whether to try to build shared libraries configure:3220: result: yes configure:3231: checking whether the compiler supports -fPIC configure:3242: distcc arm-linux-gnu-gcc -c -fPIC -DPIC -g -O2 - Werror conftest.c >&5 configure:3248: $? = 0 configure:3262: result: yes configure:3268: checking whether the compiler can link a dynamic library with -shared configure:3280: distcc arm-linux-gnu-gcc -o conftest -fPIC -DPIC -g - O2 -Werror -shared conftest.c >&5 configure:3286: $? = 0 configure:3302: result: yes configure:3310: checking extension for dynamic libraries configure:3322: result: so configure:3325: checking if the compiler supports -Wl,--rpath=. configure:3337: distcc arm-linux-gnu-gcc -o conftest -fPIC -DPIC -g - O2 -Werror -shared -Wl,-rpath=/foo conftest.c >&5 configure:3343: $? = 0 configure:3359: result: yes configure:3413: checking for grep that handles long lines and -e configure:3487: result: /bin/grep configure:3492: checking for egrep configure:3570: result: /bin/grep -E configure:3576: checking for a C compiler (trying distcc arm-linux- gnu-gcc) configure:3578: result: configure:3697: checking for C compiler version configure:3704: distcc arm-linux-gnu-gcc --version >&5 arm-linux-gnu-gcc (GCC) 4.1.2 20061115 (prerelease) (Debian 4.1.1-21) Copyright (C) 2006 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. configure:3707: $? = 0 configure:3714: distcc arm-linux-gnu-gcc -v >&5 Using built-in specs. Target: arm-linux-gnu Configured with: ../src/configure -v --enable-languages=c,c+ +,fortran,objc,obj-c++,treelang --prefix=/usr --enable-shared --with- system-zlib --libexecdir=/usr/lib --without-included-gettext --enable- threads=posix --enable-nls --program-suffix=-4.1 --enable- __cxa_atexit --enable-clocale=gnu --enable-libstdcxx-debug --enable- mpfr --enable-checking=release arm-linux-gnu Thread model: posix gcc version 4.1.2 20061115 (prerelease) (Debian 4.1.1-21) configure:3717: $? = 0 configure:3724: distcc arm-linux-gnu-gcc -V >&5 arm-linux-gnu-gcc: '-V' option must have argument distcc[26235] ERROR: compile (null) on localhost failed configure:3727: $? = 1 configure:3730: checking whether we are using the GNU C compiler configure:3782: result: yes configure:3787: checking whether distcc arm-linux-gnu-gcc accepts -g configure:3922: result: yes configure:3939: checking for distcc arm-linux-gnu-gcc option to accept ISO C89 configure:4042: result: none needed configure:4060: checking for a C++ compiler (trying dist++ arm-linux- gnu-gcc) configure:4062: result: configure:4180: checking for C++ compiler version configure:4187: dist++ arm-linux-gnu-gcc --version >&5 ./configure: line 4188: dist++: command not found configure:4190: $? = 127 configure:4197: dist++ arm-linux-gnu-gcc -v >&5 ./configure: line 4198: dist++: command not found configure:4200: $? = 127 configure:4207: dist++ arm-linux-gnu-gcc -V >&5 ./configure: line 4208: dist++: command not found configure:4210: $? = 127 configure:4213: checking whether we are using the GNU C++ compiler configure:4242: dist++ arm-linux-gnu-gcc -c conftest.cpp >&5 ./configure: line 4243: dist++: command not found configure:4248: $? = 127 configure: failed program was: | /* confdefs.h. */ | #define PACKAGE_NAME "" | #define PACKAGE_TARNAME "" | #define PACKAGE_VERSION "" | #define PACKAGE_STRING "" | #define PACKAGE_BUGREPORT "" | #define HAVE_CONFIG_STATE 1 | #define INSTALL_SYSCONFDIR "/etc" | #define INSTALL_LOCALSTATEDIR "/var" | #define OASYS_DEBUG_LOCKING_ENABLED 1 | /* end confdefs.h. */ | | int | main () | { | #ifndef __GNUC__ | choke me | #endif | | ; | return 0; | } configure:4265: result: no configure:4270: checking whether dist++ arm-linux-gnu-gcc accepts -g configure:4300: dist++ arm-linux-gnu-gcc -c -g conftest.cpp >&5 ./configure: line 4301: dist++: command not found configure:4306: $? = 127 configure: failed program was: | /* confdefs.h. */ Received: from gateway0.EECS.Berkeley.EDU (gateway0.EECS.Berkeley.EDU [169.229.60.93]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l7JHfXC02294 for ; Sun, 19 Aug 2007 10:41:33 -0700 Received: from [192.168.1.2] (dsl081-061-178.sfo1.dsl.speakeasy.net [64.81.61.178]) (authenticated bits=0) by gateway0.EECS.Berkeley.EDU (8.14.1/8.13.5) with ESMTP id l7JIfVbx014784 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Sun, 19 Aug 2007 11:41:33 -0700 (PDT) In-Reply-To: <9CE6073D-FDD4-404E-BCF9-A0967BA754FE@virgin.net> References: <9CE6073D-FDD4-404E-BCF9-A0967BA754FE@virgin.net> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: Content-Transfer-Encoding: 7bit From: Michael Demmer Subject: Re: [dtn-users] DTN2 configure problems with BerkeleyDB using distcc X-Applemailsentby: demmer Date: Sun, 19 Aug 2007 11:41:34 -0700 To: dtn-users@mailman.dtnrg.org X-Mailer: Apple Mail (2.752.3) Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: Try running configure --with-cc="distcc arm-linux-gnu-gcc" and see if that works. I don't think that the DTN2 configure actually respects the CC environment variable -- though it probably should to be consistent with other configure scripts. -m On Aug 19, 2007, at 6:25 AM, Darren Long wrote: > Hi, > > I decided to do something about the horribly slow build times on my > NSLU2 (about 12 hours for DTN2), so I got crosstool to build an arm- > linux-gnu toolchain and set up distcc so I could farm off cross- > compilation to my dual-core Ubuntu box. > > I did this following the guide at http://www.mythtv.org/wiki/ > index.php/Myth_On_NSLU2 > > However, when using distcc, I can't get configure to configure DTN2 > (fresh CVS) properly for BerkeleyDB. To use distcc, I'm exporting > the following before running configure: > > export CC="distcc arm-linux-gnu-gcc" > export CXX="distcc arm-linux-gnu-g++" > export DISTCC_HOSTS="192.168.0.4" > > I even did a fresh source build of BerkeleyDB, using distcc into / > usr/local/BerkeleyDB.4.5, so I know configuring and building with > distcc works fine. > > I've just tried a workaround, where I configure without having > exported CC and CXX as distcc, then edited the Rules.make files to > switch CC and CXX to point at distcc, then export as above, and > finally make. This seems to be working and although it's not > finished building yet, it seems much faster. > > So, in summary, DTN2 configures fine when I'm not using distcc, > both with and without the explicit --with-db path, picking up the > newly built 4.5 version, but fails to configure if I export CC and > CXX as distcc beforehand. > > Any ideas what the problem might be? > > Cheers, > > Darren > _______________________________________________ > dtn-users mailing list > dtn-users@mailman.dtnrg.org > http://mailman.dtnrg.org/mailman/listinfo/dtn-users Received: from n068.sc1.he.tucows.com (smtpout0172.sc1.he.tucows.com [64.97.136.172]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l7JFQ3C01503 for ; Sun, 19 Aug 2007 08:26:03 -0700 Received: from [192.168.0.6] (86.31.126.90) by n068.sc1.he.tucows.com (7.2.069.1) id 46C8394F00002F07 for dtn-users@mailman.dtnrg.org; Sun, 19 Aug 2007 16:25:57 +0000 Mime-Version: 1.0 (Apple Message framework v752.2) In-Reply-To: <9CE6073D-FDD4-404E-BCF9-A0967BA754FE@virgin.net> References: <9CE6073D-FDD4-404E-BCF9-A0967BA754FE@virgin.net> Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: Content-Transfer-Encoding: 7bit From: Darren Long Subject: Re: [dtn-users] DTN2 configure problems with BerkeleyDB using distcc Date: Sun, 19 Aug 2007 17:25:54 +0100 To: dtn-users@mailman.dtnrg.org X-Mailer: Apple Mail (2.752.2) Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: By the way, using distcc trimmed more than 10 hours off the build time :-) Linking is still like watching paint dry :-( It would be good to get the distcc configure problem sorted. Cheers, Darren Received: from n064.sc1.he.tucows.com (smtpout0163.sc1.he.tucows.com [64.97.136.163]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l7JCQ6C32696 for ; Sun, 19 Aug 2007 05:26:06 -0700 Received: from [192.168.0.6] (86.31.126.90) by n064.sc1.he.tucows.com (7.2.069.1) id 466CB1C80051D84A for dtn-users@mailman.dtnrg.org; Sun, 19 Aug 2007 13:26:00 +0000 Mime-Version: 1.0 (Apple Message framework v752.2) Content-Transfer-Encoding: 7bit Message-Id: <9CE6073D-FDD4-404E-BCF9-A0967BA754FE@virgin.net> Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed To: dtn-users@mailman.dtnrg.org From: Darren Long Date: Sun, 19 Aug 2007 14:25:55 +0100 X-Mailer: Apple Mail (2.752.2) Subject: [dtn-users] DTN2 configure problems with BerkeleyDB using distcc Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: Hi, I decided to do something about the horribly slow build times on my NSLU2 (about 12 hours for DTN2), so I got crosstool to build an arm- linux-gnu toolchain and set up distcc so I could farm off cross- compilation to my dual-core Ubuntu box. I did this following the guide at http://www.mythtv.org/wiki/ index.php/Myth_On_NSLU2 However, when using distcc, I can't get configure to configure DTN2 (fresh CVS) properly for BerkeleyDB. To use distcc, I'm exporting the following before running configure: export CC="distcc arm-linux-gnu-gcc" export CXX="distcc arm-linux-gnu-g++" export DISTCC_HOSTS="192.168.0.4" I even did a fresh source build of BerkeleyDB, using distcc into /usr/ local/BerkeleyDB.4.5, so I know configuring and building with distcc works fine. I've just tried a workaround, where I configure without having exported CC and CXX as distcc, then edited the Rules.make files to switch CC and CXX to point at distcc, then export as above, and finally make. This seems to be working and although it's not finished building yet, it seems much faster. So, in summary, DTN2 configures fine when I'm not using distcc, both with and without the explicit --with-db path, picking up the newly built 4.5 version, but fails to configure if I export CC and CXX as distcc beforehand. Any ideas what the problem might be? Cheers, Darren Received: from gateway0.EECS.Berkeley.EDU (gateway0.EECS.Berkeley.EDU [169.229.60.93]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l7FEMbC06705 for ; Wed, 15 Aug 2007 07:22:37 -0700 Received: from [10.212.2.71] (bldmz-nat-161-182.berkeley.intel-research.net [12.155.161.182]) (authenticated bits=0) by gateway0.EECS.Berkeley.EDU (8.14.1/8.13.5) with ESMTP id l7FFMaNO002311 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Wed, 15 Aug 2007 08:22:37 -0700 (PDT) In-Reply-To: <52950c3f0708150721v561c3071t16ce6874cb73fe2d@mail.gmail.com> References: <52950c3f0708141326v4f7ec5d6ie6588ed222ab7d0@mail.gmail.com> <52950c3f0708150721v561c3071t16ce6874cb73fe2d@mail.gmail.com> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: multipart/alternative; boundary=Apple-Mail-1-578244764 Message-Id: <890FE8E0-AD2A-4767-829A-FC3669CCC37B@cs.berkeley.edu> From: Michael Demmer Subject: Re: [dtn-users] dtn on arm X-Applemailsentby: demmer Date: Wed, 15 Aug 2007 08:22:38 -0700 To: dtn-users@mailman.dtnrg.org X-Mailer: Apple Mail (2.752.3) Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: --Apple-Mail-1-578244764 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed No need for a patch -- I committed the fixes. Let me know if I missed anything. -m On Aug 15, 2007, at 7:21 AM, Timur Alperovich wrote: > Right, that works :) > That was the first solution I was thinking of. Do you want patches > for that? It's a minor fix really. > > Timur > > On 8/14/07, Michael Demmer wrote: > > Can you try changing the "char c" declaration into "int c" and see > if that fixes the problem? getopt is supposed to return an int of > -1 when reaching the end of the option string, so my guess is that > the incompatibility occurs when casting that int to a char. > > -m > > On Aug 14, 2007, at 1:26 PM, Timur Alperovich wrote: > >> Hi, >> >> I've been putting dtn on small nodes, gumstix (www.gumstix.com). >> What I noticed is that I could not run the dtnsend, dtnrecv >> commands as well as a few others. The commands would simply print >> out the usage information regardless of the arguments and no >> errors. It turned out in the code for these there is an assignment >> such as "char c = getopt(...);", which results in c = 255 when >> getopt returns -1 on completion on the arm7. I found there are two >> ways to avoid this: 1. use code similar to that dtnperf: int c; >> while((c = getopt(...)) != EOF); 2. add case 255 in the switch >> statement that also results in the "done" condition. I think the >> first approach is preferred. Anyway, I wanted to point this out >> for anyone who is doing anything similar. I can submit patches for >> files to change the parsing if that would be useful. >> >> Timur > > --Apple-Mail-1-578244764 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=US-ASCII No need for a patch -- I committed the fixes. Let me know if I missed = anything.

-m

O= n Aug 15, 2007, at 7:21 AM, Timur Alperovich wrote:

Right, = that works :)
That was the first solution I was thinking of. Do you = want patches for that? It's a minor fix = really.

Timur

On 8/14/07, = Michael Demmer <demmer@cs.berkeley.edu> = wrote:

Can you try changing the = "char c" declaration into "int c" and see if that fixes the problem? = getopt is supposed to return an int of -1 when reaching the end of the = option string, so my guess is that the incompatibility occurs when = casting that int to a char.

-m

On Aug 14, 2007, at 1:26 = PM, Timur Alperovich wrote:

Hi,

= I've been putting dtn on small nodes, gumstix (www.gumstix.com). What I = noticed is that I could not run the dtnsend, dtnrecv commands as well as = a few others. The commands would simply print out the usage information = regardless of the arguments and no errors. It turned out in the code for = these there is an assignment such as "char c =3D getopt(...);", which = results in c =3D 255 when getopt returns -1 on completion on the arm7. I = found there are two ways to avoid this: 1. use code similar to that = dtnperf: int c; while((c =3D getopt(...)) !=3D EOF); 2. add case 255 in = the switch statement that also results in the "done" condition. I think = the first approach is preferred. Anyway, I wanted to point this out for = anyone who is doing anything similar. I can submit patches for files to = change the parsing if that would be useful. =

Timur



= --Apple-Mail-1-578244764-- Received: from rv-out-0910.google.com (rv-out-0910.google.com [209.85.198.189]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l7FDLLC06256 for ; Wed, 15 Aug 2007 06:21:21 -0700 Received: by rv-out-0910.google.com with SMTP id l15so157386rvb for ; Wed, 15 Aug 2007 07:21:21 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=lXXBMuDLczE7JLtvhahdASbAM/estIRDcnZUAhXP2tYYotLQ2rZUArLA1dfaYVZs+PJawjdDOE7PvdkyVybsJIvcyNuuTfQToSnTBW/pJv9ecFYkygQdcGgOPSj6rTzkVB/pu7BAofZtfZyoyCqIQoRdBqfaK3VBn0GcLSCiMm0= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=P1LVPHB0KIemram3K3WELkhJXcIcsWPiDz7HkoJd5kM3QBW6LDp0oorZjX+DofIBYTRreYkKZopKMLkqVrmpFysp15UZ2q/cd8nAVcJ3lh6Y4LkF2RFk0pAGB9Q2vxSsb7MbQ9GmZp9LdLshUsEzysEJL0u/Y1W653PlPmP4MCs= Received: by 10.141.88.3 with SMTP id q3mr193902rvl.1187187680914; Wed, 15 Aug 2007 07:21:20 -0700 (PDT) Received: by 10.141.164.16 with HTTP; Wed, 15 Aug 2007 07:21:20 -0700 (PDT) Message-ID: <52950c3f0708150721v561c3071t16ce6874cb73fe2d@mail.gmail.com> Date: Wed, 15 Aug 2007 10:21:20 -0400 From: "Timur Alperovich" To: dtn-users@mailman.dtnrg.org Subject: Re: [dtn-users] dtn on arm In-Reply-To: MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_20334_3968903.1187187680891" References: <52950c3f0708141326v4f7ec5d6ie6588ed222ab7d0@mail.gmail.com> Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: ------=_Part_20334_3968903.1187187680891 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Right, that works :) That was the first solution I was thinking of. Do you want patches for that? It's a minor fix really. Timur On 8/14/07, Michael Demmer wrote: > > > Can you try changing the "char c" declaration into "int c" and see if that > fixes the problem? getopt is supposed to return an int of -1 when reaching > the end of the option string, so my guess is that the incompatibility occurs > when casting that int to a char. > > -m > > On Aug 14, 2007, at 1:26 PM, Timur Alperovich wrote: > > Hi, > > I've been putting dtn on small nodes, gumstix (www.gumstix.com). What I > noticed is that I could not run the dtnsend, dtnrecv commands as well as a > few others. The commands would simply print out the usage information > regardless of the arguments and no errors. It turned out in the code for > these there is an assignment such as "char c = getopt(...);", which results > in c = 255 when getopt returns -1 on completion on the arm7. I found there > are two ways to avoid this: 1. use code similar to that dtnperf: int c; > while((c = getopt(...)) != EOF); 2. add case 255 in the switch statement > that also results in the "done" condition. I think the first approach is > preferred. Anyway, I wanted to point this out for anyone who is doing > anything similar. I can submit patches for files to change the parsing if > that would be useful. > > Timur > > > ------=_Part_20334_3968903.1187187680891 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Right, that works :)
That was the first solution I was thinking of. Do you want patches for that? It's a minor fix really.

Timur

On 8/14/07, Michael Demmer <demmer@cs.berkeley.edu> wrote:

Can you try changing the "char c" declaration into "int c" and see if that fixes the problem? getopt is supposed to return an int of -1 when reaching the end of the option string, so my guess is that the incompatibility occurs when casting that int to a char.

-m

On Aug 14, 2007, at 1:26 PM, Timur Alperovich wrote:

Hi,

I've been putting dtn on small nodes, gumstix (www.gumstix.com). What I noticed is that I could not run the dtnsend, dtnrecv commands as well as a few others. The commands would simply print out the usage information regardless of the arguments and no errors. It turned out in the code for these there is an assignment such as "char c = getopt(...);", which results in c = 255 when getopt returns -1 on completion on the arm7. I found there are two ways to avoid this: 1. use code similar to that dtnperf: int c; while((c = getopt(...)) != EOF); 2. add case 255 in the switch statement that also results in the "done" condition. I think the first approach is preferred. Anyway, I wanted to point this out for anyone who is doing anything similar. I can submit patches for files to change the parsing if that would be useful.

Timur


------=_Part_20334_3968903.1187187680891-- Received: from gateway0.EECS.Berkeley.EDU (gateway0.EECS.Berkeley.EDU [169.229.60.93]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l7EJdOC24230 for ; Tue, 14 Aug 2007 12:39:24 -0700 Received: from [10.212.2.71] (bldmz-nat-161-182.berkeley.intel-research.net [12.155.161.182]) (authenticated bits=0) by gateway0.EECS.Berkeley.EDU (8.14.1/8.13.5) with ESMTP id l7EKdN0e023728 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Tue, 14 Aug 2007 13:39:24 -0700 (PDT) In-Reply-To: <52950c3f0708141326v4f7ec5d6ie6588ed222ab7d0@mail.gmail.com> References: <52950c3f0708141326v4f7ec5d6ie6588ed222ab7d0@mail.gmail.com> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: multipart/alternative; boundary=Apple-Mail-2-510850283 Message-Id: From: Michael Demmer Subject: Re: [dtn-users] dtn on arm X-Applemailsentby: demmer Date: Tue, 14 Aug 2007 13:39:24 -0700 To: dtn-users@mailman.dtnrg.org X-Mailer: Apple Mail (2.752.3) Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: --Apple-Mail-2-510850283 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Can you try changing the "char c" declaration into "int c" and see if that fixes the problem? getopt is supposed to return an int of -1 when reaching the end of the option string, so my guess is that the incompatibility occurs when casting that int to a char. -m On Aug 14, 2007, at 1:26 PM, Timur Alperovich wrote: > Hi, > > I've been putting dtn on small nodes, gumstix (www.gumstix.com). > What I noticed is that I could not run the dtnsend, dtnrecv > commands as well as a few others. The commands would simply print > out the usage information regardless of the arguments and no > errors. It turned out in the code for these there is an assignment > such as "char c = getopt(...);", which results in c = 255 when > getopt returns -1 on completion on the arm7. I found there are two > ways to avoid this: 1. use code similar to that dtnperf: int c; > while((c = getopt(...)) != EOF); 2. add case 255 in the switch > statement that also results in the "done" condition. I think the > first approach is preferred. Anyway, I wanted to point this out for > anyone who is doing anything similar. I can submit patches for > files to change the parsing if that would be useful. > > Timur --Apple-Mail-2-510850283 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=US-ASCII

Can you try = changing the "char c" declaration into "int c" and see if that fixes the = problem? getopt is supposed to return an int of -1 when reaching the end = of the option string, so my guess is that the incompatibility occurs = when casting that int to a char.

-m

O= n Aug 14, 2007, at 1:26 PM, Timur Alperovich wrote:

Hi,

I've been putting dtn on small nodes, gumstix = (www.gumstix.com). What I noticed = is that I could not run the dtnsend, dtnrecv commands as well as a few = others. The commands would simply print out the usage information = regardless of the arguments and no errors. It turned out in the code for = these there is an assignment such as "char c =3D getopt(...);", which = results in c =3D 255 when getopt returns -1 on completion on the arm7. I = found there are two ways to avoid this: 1. use code similar to that = dtnperf: int c; while((c =3D getopt(...)) !=3D EOF); 2. add case 255 in = the switch statement that also results in the "done" condition. I think = the first approach is preferred. Anyway, I wanted to point this out for = anyone who is doing anything similar. I can submit patches for files to = change the parsing if that would be useful. =

Timur

= --Apple-Mail-2-510850283-- Received: from rv-out-0910.google.com (rv-out-0910.google.com [209.85.198.191]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l7EJQJC24144 for ; Tue, 14 Aug 2007 12:26:19 -0700 Received: by rv-out-0910.google.com with SMTP id f1so1385789rvb for ; Tue, 14 Aug 2007 13:26:19 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:mime-version:content-type; b=fl+CB9rLii/UNMX5zByZOm0Viff2fGOwZkmQenjoFWsL7qZODCbFOUb8J17S8fCav98wotrfiBcHC2piIhdMwxdrfWg1NVB5o6DytL+wpimgXhQVC4foblyL96lUBUv1Ddl6eyzAOOS4gV4pjKT121S7FWQGPbfBYckXU1iuCis= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:mime-version:content-type; b=WkLfqS6HlNvwZa6DBa/paT068A34+zKuRQFURzWoKTEt7W37Mkpj6hWY8/6P8DF9M9diXm4l/Buk6qrzkLcomskt5q/3jE46fV5X20lC5wDsffVX2BATruQxE6rKITDnt2KtvhC4tj5tVH3m0IXF7J75hAD4Ae+bUPaSCO6KkFM= Received: by 10.141.76.21 with SMTP id d21mr2825377rvl.1187123178542; Tue, 14 Aug 2007 13:26:18 -0700 (PDT) Received: by 10.141.164.16 with HTTP; Tue, 14 Aug 2007 13:26:18 -0700 (PDT) Message-ID: <52950c3f0708141326v4f7ec5d6ie6588ed222ab7d0@mail.gmail.com> Date: Tue, 14 Aug 2007 16:26:18 -0400 From: "Timur Alperovich" To: dtn-users@mailman.dtnrg.org MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_18000_14281115.1187123178517" Subject: [dtn-users] dtn on arm Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: ------=_Part_18000_14281115.1187123178517 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Hi, I've been putting dtn on small nodes, gumstix (www.gumstix.com). What I noticed is that I could not run the dtnsend, dtnrecv commands as well as a few others. The commands would simply print out the usage information regardless of the arguments and no errors. It turned out in the code for these there is an assignment such as "char c = getopt(...);", which results in c = 255 when getopt returns -1 on completion on the arm7. I found there are two ways to avoid this: 1. use code similar to that dtnperf: int c; while((c = getopt(...)) != EOF); 2. add case 255 in the switch statement that also results in the "done" condition. I think the first approach is preferred. Anyway, I wanted to point this out for anyone who is doing anything similar. I can submit patches for files to change the parsing if that would be useful. Timur ------=_Part_18000_14281115.1187123178517 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Hi,

I've been putting dtn on small nodes, gumstix (www.gumstix.com). What I noticed is that I could not run the dtnsend, dtnrecv commands as well as a few others. The commands would simply print out the usage information regardless of the arguments and no errors. It turned out in the code for these there is an assignment such as "char c = getopt(...);", which results in c = 255 when getopt returns -1 on completion on the arm7. I found there are two ways to avoid this: 1. use code similar to that dtnperf: int c; while((c = getopt(...)) != EOF); 2. add case 255 in the switch statement that also results in the "done" condition. I think the first approach is preferred. Anyway, I wanted to point this out for anyone who is doing anything similar. I can submit patches for files to change the parsing if that would be useful.

Timur
------=_Part_18000_14281115.1187123178517-- Received: from rv-out-0910.google.com (rv-out-0910.google.com [209.85.198.189]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l7DHXvC03465 for ; Mon, 13 Aug 2007 10:33:57 -0700 Received: by rv-out-0910.google.com with SMTP id f1so1130263rvb for ; Mon, 13 Aug 2007 11:33:57 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=n53cwJmd9QxnD1+BYDf5wBB+541Z0SvEgvsVut4/r2azCG6bYIuz7+DSPOR3qfSIfJuCt70u3GGVSDYi7Dm16c7iOY+jW5SBvPgPiD4eQCAijwh96Omj5JEH9APNqW5u+i4eX2TKOn2YlasyfQZa9x+jf5kI3ibzRPB1+fDYDCQ= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=OQZu+JM2s9+kU0VrIEZqU6BMyouZ75eZ2Qc9kRyKIjmIkJJ3SOw8DktXU3rlI3cfiiK4KncSpC4muu7vQ1HidFrUo3huuaj2ZROkHwBIPkn9gCpN46W/TWbglDoHV3WcGp06HfMs5+DlEC+mQsDOtsExwumPfZPNKlauqj1Wjto= Received: by 10.141.71.19 with SMTP id y19mr2670995rvk.1187030036897; Mon, 13 Aug 2007 11:33:56 -0700 (PDT) Received: by 10.141.164.16 with HTTP; Mon, 13 Aug 2007 11:33:56 -0700 (PDT) Message-ID: <52950c3f0708131133p320a27e2w1cd8b96ded0e34c9@mail.gmail.com> Date: Mon, 13 Aug 2007 14:33:56 -0400 From: "Timur Alperovich" To: dtn-users@mailman.dtnrg.org Subject: Re: [dtn-users] more multi hop issues In-Reply-To: <1C6492E9-1FDF-4701-954D-C859BDDB3416@cs.berkeley.edu> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_13740_9975667.1187030036884" References: <52950c3f0708091236q7961d778gc77e84e63882b350@mail.gmail.com> <1C6492E9-1FDF-4701-954D-C859BDDB3416@cs.berkeley.edu> Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: ------=_Part_13740_9975667.1187030036884 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Michael, Thanks for the help. I think it works now. I was able to send a dtn message from one node to another over one hop. Timur On 8/9/07, Michael Demmer wrote: > > I just committed a change to the cvs HEAD so the daemon will > typically by default open opportunistic links that are found by the > discovery mechanism. let me know if that works better for you. > > -m > > > On Aug 9, 2007, at 12:36 PM, Timur Alperovich wrote: > > > Hi, > > > > I apologize for the stream of questions but I seem to be stuck > > again. I have discoveries being sent on the multi-homed nodes and > > nodes appear to detect their neighbors, but there are no routes > > added for them and no multi-hop neighbors are added. What could be > > the reason for this? I'm just trying to get a simple multi-hop > > network to work. > > > > Timur > > _______________________________________________ > dtn-users mailing list > dtn-users@mailman.dtnrg.org > http://mailman.dtnrg.org/mailman/listinfo/dtn-users > ------=_Part_13740_9975667.1187030036884 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Michael,

Thanks for the help. I think it works now. I was able to send a dtn message from one node to another over one hop.

Timur

On 8/9/07, Michael Demmer <demmer@cs.berkeley.edu> wrote:
I just committed a change to the cvs HEAD so the daemon will
typically by default open opportunistic links that are found by the
discovery mechanism. let me know if that works better for you.

-m


On Aug 9, 2007, at 12:36 PM, Timur Alperovich wrote:

> Hi,
>
> I apologize for the stream of questions but I seem to be stuck
> again. I have discoveries being sent on the multi-homed nodes and
> nodes appear to detect their neighbors, but there are no routes
> added for them and no multi-hop neighbors are added. What could be
> the reason for this? I'm just trying to get a simple multi-hop
> network to work.
>
> Timur

_______________________________________________
dtn-users mailing list
dtn-users@mailman.dtnrg.org
http://mailman.dtnrg.org/mailman/listinfo/dtn-users

------=_Part_13740_9975667.1187030036884-- Received: from gateway0.EECS.Berkeley.EDU (gateway0.EECS.Berkeley.EDU [169.229.60.93]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l79N9kC18734 for ; Thu, 9 Aug 2007 16:09:46 -0700 Received: from [10.212.2.161] (bldmz-nat-161-182.berkeley.intel-research.net [12.155.161.182]) (authenticated bits=0) by gateway0.EECS.Berkeley.EDU (8.14.1/8.13.5) with ESMTP id l7A09fch010520 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Thu, 9 Aug 2007 17:09:46 -0700 (PDT) In-Reply-To: <52950c3f0708091236q7961d778gc77e84e63882b350@mail.gmail.com> References: <52950c3f0708091236q7961d778gc77e84e63882b350@mail.gmail.com> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: <1C6492E9-1FDF-4701-954D-C859BDDB3416@cs.berkeley.edu> Content-Transfer-Encoding: 7bit From: Michael Demmer Subject: Re: [dtn-users] more multi hop issues X-Applemailsentby: demmer Date: Thu, 9 Aug 2007 17:09:40 -0700 To: dtn-users@mailman.dtnrg.org X-Mailer: Apple Mail (2.752.3) Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: I just committed a change to the cvs HEAD so the daemon will typically by default open opportunistic links that are found by the discovery mechanism. let me know if that works better for you. -m On Aug 9, 2007, at 12:36 PM, Timur Alperovich wrote: > Hi, > > I apologize for the stream of questions but I seem to be stuck > again. I have discoveries being sent on the multi-homed nodes and > nodes appear to detect their neighbors, but there are no routes > added for them and no multi-hop neighbors are added. What could be > the reason for this? I'm just trying to get a simple multi-hop > network to work. > > Timur Received: from gateway0.EECS.Berkeley.EDU (gateway0.EECS.Berkeley.EDU [169.229.60.93]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l79JKvC17295 for ; Thu, 9 Aug 2007 12:20:57 -0700 Received: from [10.212.2.161] (bldmz-nat-161-182.berkeley.intel-research.net [12.155.161.182]) (authenticated bits=0) by gateway0.EECS.Berkeley.EDU (8.14.1/8.13.5) with ESMTP id l79KKu0s006808 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Thu, 9 Aug 2007 13:20:57 -0700 (PDT) In-Reply-To: <52950c3f0708091236q7961d778gc77e84e63882b350@mail.gmail.com> References: <52950c3f0708091236q7961d778gc77e84e63882b350@mail.gmail.com> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: <15CD9019-4E9A-4786-8B92-3CD69FCCD8C3@cs.berkeley.edu> Content-Transfer-Encoding: 7bit From: Michael Demmer Subject: Re: [dtn-users] more multi hop issues X-Applemailsentby: demmer Date: Thu, 9 Aug 2007 13:20:54 -0700 To: dtn-users@mailman.dtnrg.org X-Mailer: Apple Mail (2.752.3) Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: I think you're hitting a flaw in the routing code (in this case DTLSR) in that it doesn't actually open the discovered links, so never finds out about other nodes that are reachable through them.... I'm going to work on a patch now, stay tuned. -m On Aug 9, 2007, at 12:36 PM, Timur Alperovich wrote: > Hi, > > I apologize for the stream of questions but I seem to be stuck > again. I have discoveries being sent on the multi-homed nodes and > nodes appear to detect their neighbors, but there are no routes > added for them and no multi-hop neighbors are added. What could be > the reason for this? I'm just trying to get a simple multi-hop > network to work. > > Timur Received: from rv-out-0910.google.com (rv-out-0910.google.com [209.85.198.185]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l79IaIC16891 for ; Thu, 9 Aug 2007 11:36:18 -0700 Received: by rv-out-0910.google.com with SMTP id f1so438845rvb for ; Thu, 09 Aug 2007 12:36:13 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:mime-version:content-type; b=eVOtzmG+CLouQBfGyYvfEs4zByNYCuNwwU8IlyyZRR617+TRkHbyERWAOcgrhWJFtV260yKSHKnVwhSEvRA0zemyJ3NwsJ2wBBxXUfAG2qVkr82OS02TBZQbigMkbUgSBPEHiszzowUkJi9G7l4iVqiisWnPGHr8OB7Yy/jLS7c= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:mime-version:content-type; b=ROgVSDs2Nhanjq9BpSLkc6zZ7BvLzU3UIqhifp20brvl1Ytt+7cDBldMSslS9JUgPAvjRnuUm50cPANLBji+/IafAWfCi0wlzLbwhtQdU4h4PNQoJV9v0ap6YuHTVXPiZyPZKe8+3OxYBGnPlzyq8it6s9Ui9bXCPKzxbjVDGrE= Received: by 10.140.191.14 with SMTP id o14mr1094215rvf.1186688172922; Thu, 09 Aug 2007 12:36:12 -0700 (PDT) Received: by 10.141.164.16 with HTTP; Thu, 9 Aug 2007 12:36:12 -0700 (PDT) Message-ID: <52950c3f0708091236q7961d778gc77e84e63882b350@mail.gmail.com> Date: Thu, 9 Aug 2007 15:36:12 -0400 From: "Timur Alperovich" To: dtn-users@mailman.dtnrg.org MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_2961_30333414.1186688172907" Subject: [dtn-users] more multi hop issues Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: ------=_Part_2961_30333414.1186688172907 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Hi, I apologize for the stream of questions but I seem to be stuck again. I have discoveries being sent on the multi-homed nodes and nodes appear to detect their neighbors, but there are no routes added for them and no multi-hop neighbors are added. What could be the reason for this? I'm just trying to get a simple multi-hop network to work. Timur ------=_Part_2961_30333414.1186688172907 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Hi,

I apologize for the stream of questions but I seem to be stuck again. I have discoveries being sent on the multi-homed nodes and nodes appear to detect their neighbors, but there are no routes added for them and no multi-hop neighbors are added. What could be the reason for this? I'm just trying to get a simple multi-hop network to work.

Timur
------=_Part_2961_30333414.1186688172907-- Received: from gateway0.EECS.Berkeley.EDU (gateway0.EECS.Berkeley.EDU [169.229.60.93]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l78GR7C04214; Wed, 8 Aug 2007 09:27:07 -0700 Received: from [192.168.1.2] (dsl081-061-178.sfo1.dsl.speakeasy.net [64.81.61.178]) (authenticated bits=0) by gateway0.EECS.Berkeley.EDU (8.14.1/8.13.5) with ESMTP id l78HR5eQ020591 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Wed, 8 Aug 2007 10:27:06 -0700 (PDT) In-Reply-To: References: Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: <6C97032C-3709-42E2-BA8E-F8E435DE1DEA@cs.berkeley.edu> Cc: dtn-users@mailman.dtnrg.org, dtn-dev Content-Transfer-Encoding: 7bit From: Michael Demmer X-Applemailsentby: demmer Date: Wed, 8 Aug 2007 10:27:03 -0700 To: "Scott, Keith L." X-Mailer: Apple Mail (2.752.3) Subject: [dtn-users] Re: [dtn-dev] changing version control systems Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: Good point -- even with the distributed VC systems, I would still maintain a master copy of the source tree on code.dtnrg.org. Anyone will be able to pull from that using http. There's also a way to push changes over ssh, just like in CVS today. The advantage is that for an individual developer, you can do incremental checkins to your own "repository". With mercurial at least, you can easily share changes with other people inside a firewall in an ad-hoc manner, since it can run in an http server mode as well. But the master copy would be controlled via ssh like it is today. -m On Aug 8, 2007, at 10:14 AM, Scott, Keith L. wrote: > I don't have any experience with subversion or Mercurial. My one > comment with respect to distributed revision control is that some > of us > spend a good portion of our lives behind rather strong firewalls where > it is virtually impossible to instantiate inbound connections to our > machines. In such a configuration, I suppose our (firewalled > peoples') > interactions would primarily be via the distributed repository process > running at code.dtnrg.org? > > --keith > > > -----Original Message----- > From: dtn-dev-admin@mailman.dtnrg.org > [mailto:dtn-dev-admin@mailman.dtnrg.org] On Behalf Of Michael Demmer > Sent: Tuesday, August 07, 2007 8:44 PM > To: dtn-users@mailman.dtnrg.org; dtn-dev > Subject: [dtn-dev] changing version control systems > > > I'm considering changing the main DTN reference implementation source > repository from CVS to subversion[1] or mercurial[2]. > > There are numerous reasons why these other systems are better version > control mechanisms than CVS, including the ability to rename files, > grouping modifications into logical changesets, directories that are > editable in the VC system, and better support for branching, to name > a few. Mercurial adds the additional advantages of being a > distributed rather than centralized system. > > On the other hand, switching will likely entail transition pain, and > some people may not be familiar with the new systems. In particular, > mercurial has a very different usage model which will change how > people interact with the source repository. > > The intent of this mail is to get a sense of the overall temperature > of the DTN user base out there with respect to this change: > > Do you think it's a great idea? Is it a terrible idea? > Will it impede your interactions with the code? > Do you have horror stories about subversion and/or mercurial? > Does anyone have a strong preference for one versus the other? > > -m > > --- > > [1] http://subversion.tigris.org/ > [2] http://www.selenic.com/mercurial/wiki/ > _______________________________________________ > dtn-dev mailing list > dtn-dev@mailman.dtnrg.org > http://mailman.dtnrg.org/mailman/listinfo/dtn-dev Received: from smtp-mclean.mitre.org (smtpproxy2.mitre.org [192.80.55.71]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l78GEmC04078; Wed, 8 Aug 2007 09:14:48 -0700 Received: from smtp-mclean.mitre.org (localhost.localdomain [127.0.0.1]) by smtp-mclean.mitre.org (8.12.11.20060308/8.12.11) with SMTP id l78HEmGc005652; Wed, 8 Aug 2007 13:14:48 -0400 Received: from smtp-mclean.mitre.org (localhost.localdomain [127.0.0.1]) by smtp-mclean.mitre.org (Postfix) with ESMTP id 127EE4F8D8; Wed, 8 Aug 2007 13:14:48 -0400 (EDT) Received: from imcfe2.MITRE.ORG (imcfe2.mitre.org [129.83.29.4]) by smtp-mclean.mitre.org (8.12.11.20060308/8.12.11) with ESMTP id l78HElWd005578; Wed, 8 Aug 2007 13:14:47 -0400 Received: from IMCSRV1.MITRE.ORG ([129.83.20.158]) by imcfe2.MITRE.ORG with Microsoft SMTPSVC(6.0.3790.1830); Wed, 8 Aug 2007 13:14:47 -0400 X-MimeOLE: Produced By Microsoft Exchange V6.5 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Wed, 8 Aug 2007 13:14:43 -0400 Message-ID: In-Reply-To: X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: [dtn-dev] changing version control systems Thread-Index: AcfZVYEMa21PJXbOQCa9i7SgpUecMwAd2XuA From: "Scott, Keith L." To: "Michael Demmer" , , "dtn-dev" X-OriginalArrivalTime: 08 Aug 2007 17:14:47.0214 (UTC) FILETIME=[9F0444E0:01C7D9DF] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by webbie.berkeley.intel-research.net id l78GEmC04078 Subject: [dtn-users] RE: [dtn-dev] changing version control systems Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: I don't have any experience with subversion or Mercurial. My one comment with respect to distributed revision control is that some of us spend a good portion of our lives behind rather strong firewalls where it is virtually impossible to instantiate inbound connections to our machines. In such a configuration, I suppose our (firewalled peoples') interactions would primarily be via the distributed repository process running at code.dtnrg.org? --keith -----Original Message----- From: dtn-dev-admin@mailman.dtnrg.org [mailto:dtn-dev-admin@mailman.dtnrg.org] On Behalf Of Michael Demmer Sent: Tuesday, August 07, 2007 8:44 PM To: dtn-users@mailman.dtnrg.org; dtn-dev Subject: [dtn-dev] changing version control systems I'm considering changing the main DTN reference implementation source repository from CVS to subversion[1] or mercurial[2]. There are numerous reasons why these other systems are better version control mechanisms than CVS, including the ability to rename files, grouping modifications into logical changesets, directories that are editable in the VC system, and better support for branching, to name a few. Mercurial adds the additional advantages of being a distributed rather than centralized system. On the other hand, switching will likely entail transition pain, and some people may not be familiar with the new systems. In particular, mercurial has a very different usage model which will change how people interact with the source repository. The intent of this mail is to get a sense of the overall temperature of the DTN user base out there with respect to this change: Do you think it's a great idea? Is it a terrible idea? Will it impede your interactions with the code? Do you have horror stories about subversion and/or mercurial? Does anyone have a strong preference for one versus the other? -m --- [1] http://subversion.tigris.org/ [2] http://www.selenic.com/mercurial/wiki/ _______________________________________________ dtn-dev mailing list dtn-dev@mailman.dtnrg.org http://mailman.dtnrg.org/mailman/listinfo/dtn-dev Received: from rv-out-0910.google.com (rv-out-0910.google.com [209.85.198.188]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l78G1HC03993 for ; Wed, 8 Aug 2007 09:01:17 -0700 Received: by rv-out-0910.google.com with SMTP id f1so165042rvb for ; Wed, 08 Aug 2007 10:01:14 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=hHvXZ+nZfWOFfBHpsxFgq19AZ/J4y0EdJh4JQVR5HwtTDjkNMicOFMDVolb9+UJUKNgFHIqbMwO/uoSQdsLAtKjnrHuzaJ9YntlBUfeAh8nPFEBoBhwbAMeOp2m5jepOI3ukc9iiU4hncZpRzQbRvTVVo/+oxKkoBoHRHeteo8w= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=WbDF7unWhBnG0gGsBFdmMwAbEqvDn0t+E71fV8mRzgecHLSyv1NOF7pDhog3eiXXZpobPy+WD0tnJMVre+iI5AUJcwsu3OVCoxWmNcWoYlNHf4w9GGKvjp3/rVCvdrWtkFIVv+Vz5g8K0wVO/1hLMcmvbBnKqlhN2sGlxfMnWTU= Received: by 10.141.209.7 with SMTP id l7mr578700rvq.1186592473911; Wed, 08 Aug 2007 10:01:13 -0700 (PDT) Received: by 10.141.51.12 with HTTP; Wed, 8 Aug 2007 10:01:13 -0700 (PDT) Message-ID: <52950c3f0708081001o21e025c5u221684e35e13e38d@mail.gmail.com> Date: Wed, 8 Aug 2007 13:01:13 -0400 From: "Timur Alperovich" To: dtn-users@mailman.dtnrg.org Subject: Re: [dtn-users] Fwd: dtn on a three node multi-hop In-Reply-To: MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_949_13989698.1186592473889" References: <52950c3f0708071606w5ec0a4efr2a93bcb6e6d8f4f1@mail.gmail.com> <52950c3f0708071611q1b293893gc45938c1a3ce6b58@mail.gmail.com> <52950c3f0708071640g756bdd49l909718ab4cb19f29@mail.gmail.com> <52950c3f0708080749j264f2f95ke37255abf03788c3@mail.gmail.com> <52950c3f0708080840uf07fc3fg2bf3842df40ab5e4@mail.gmail.com> Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: ------=_Part_949_13989698.1186592473889 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline I actually got it to work now: I had to set the local_addr of each of the interfaces when creating the dtn interfaces and then use those to set up discovery. It seems to be working now. Timur On 8/8/07, Michael Demmer wrote: > > > The interfaces are only related to incoming bundles, and so yes, only > impact which address the node listens on for connections. > > You might want to try the bonjour discovery -- If you're on Linux, you can > install the avahi-daemon to get the mDNS stack. The ip discovery hasn't been > well-tested on multi-homed systems (as you've seen) so it may not set up the > bindings properly. No guarantees that bonjour will do better, but it's worth > a shot. > > -m > > On Aug 8, 2007, at 8:40 AM, Timur Alperovich wrote: > > Thanks. I'll give that a try. What about specifying local_addr in the > interface command? Does that only set the listening address? > > I'm using the ip module right now. From what I read in the help it appears > only bluetooth and ip exist. Are there any others which may be better suited > for what I'm doing? > > Timur > > On 8/8/07, Michael Demmer wrote: > > > > There's a local_addr option in the link configuration which tells it > > which interface to use, if you were to set up an explicit link to the > > other node. For example, adding this to node B's config: > > > > link add l-C 10.0.0.C ALWAYSON TCP local_addr=10.0.0.B > > > > This assumes that 10.0.0.C is C's ip address on the ad-hoc network > > and 10.0.0.B is B's address on the ad-hoc network. The local_addr > > option tells the node which address to bind to before connecting to > > the other node. > > > > Given that, I'd love it if the discovery mechanism were to figure out > > the situation automatically. Which discovery module are you using? > > > > -m > > > ------=_Part_949_13989698.1186592473889 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline I actually got it to work now: I had to set the local_addr of each of the interfaces when creating the dtn interfaces and then use those to set up discovery. It seems to be working now.

Timur

On 8/8/07, Michael Demmer <demmer@cs.berkeley.edu> wrote:

The interfaces are only related to incoming bundles, and so yes, only impact which address the node listens on for connections.

You might want to try the bonjour discovery -- If you're on Linux, you can install the avahi-daemon to get the mDNS stack. The ip discovery hasn't been well-tested on multi-homed systems (as you've seen) so it may not set up the bindings properly. No guarantees that bonjour will do better, but it's worth a shot.

-m

On Aug 8, 2007, at 8:40 AM, Timur Alperovich wrote:

Thanks. I'll give that a try. What about specifying local_addr in the interface command? Does that only set the listening address?

I'm using the ip module right now. From what I read in the help it appears only bluetooth and ip exist. Are there any others which may be better suited for what I'm doing?

Timur

On 8/8/07, Michael Demmer <demmer@cs.berkeley.edu> wrote:
There's a local_addr option in the link configuration which tells it
which interface to use, if you were to set up an explicit link to the
other node. For example, adding this to node B's config:

link add l-C 10.0.0.C ALWAYSON TCP local_addr=10.0.0.B

This assumes that 10.0.0.C is C's ip address on the ad-hoc network
and 10.0.0.B is B's address on the ad-hoc network. The local_addr
option tells the node which address to bind to before connecting to
the other node.

Given that, I'd love it if the discovery mechanism were to figure out
the situation automatically. Which discovery module are you using?

-m
------=_Part_949_13989698.1186592473889-- Received: from rv-out-0910.google.com (rv-out-0910.google.com [209.85.198.191]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l78FxVC03974 for ; Wed, 8 Aug 2007 08:59:31 -0700 Received: by rv-out-0910.google.com with SMTP id f1so164662rvb for ; Wed, 08 Aug 2007 09:59:26 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:sender:to:subject:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references:x-google-sender-auth; b=EocMX//39MF6v+nZbQgBpCUlVQbdD4QfIU0rf1Fj2xmwSfTuGyyd57+JmFqHKpjaegsLXvoQgmbeb9qq767mlT/Fpjp3ziJzgVcoDtXw0evs/tvhGcRELshlcE4pPmVdj/8sHn0WN7SGtHq5jNogF8voO/vjaR7erbxws0OIUQw= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:sender:to:subject:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references:x-google-sender-auth; b=BXVPvcaDbaJrEI/BRZKGgtX87rV4p00Uxgr1YxQJpgPxqL+zV3XEOeYEF/Jw82vujTTnekvSIN5NxYTrD4UPa6ppHAfULpiuv6pYt9NB0VrdM5beGLEK88mA4QOawoNXe0s9sxl5UReIER+szq07xGjBufH+eV88wKX3nVHRe7M= Received: by 10.115.107.1 with SMTP id j1mr1523866wam.1186592365345; Wed, 08 Aug 2007 09:59:25 -0700 (PDT) Received: by 10.114.107.5 with HTTP; Wed, 8 Aug 2007 09:59:25 -0700 (PDT) Message-ID: <5f30e2610708080959i613737c7q4354e67434d9632f@mail.gmail.com> Date: Wed, 8 Aug 2007 11:59:25 -0500 From: "Jeff Wilson" To: dtn-users@mailman.dtnrg.org Subject: Re: [dtn-users] Fwd: dtn on a three node multi-hop In-Reply-To: <479F6932-2CA1-4865-909C-B21F01B7E539@cs.baylor.edu> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <52950c3f0708071606w5ec0a4efr2a93bcb6e6d8f4f1@mail.gmail.com> <52950c3f0708071611q1b293893gc45938c1a3ce6b58@mail.gmail.com> <52950c3f0708071640g756bdd49l909718ab4cb19f29@mail.gmail.com> <52950c3f0708080749j264f2f95ke37255abf03788c3@mail.gmail.com> <52950c3f0708080840uf07fc3fg2bf3842df40ab5e4@mail.gmail.com> <479F6932-2CA1-4865-909C-B21F01B7E539@cs.baylor.edu> X-Google-Sender-Auth: f35ac0aa80fa0eed Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: This weakness on multi-homed systems may well be inherent to the design of the discovery module. Both the IP and the Bluetooth modules were designed so that the discovery piece applies to the address family, and the announce piece serves as the registration for each distinct Interface (listening Convergence Layer instance). With hindsight, it seems like a better approach is to move the discovery module to represent a physical interface, and the announce piece remains as a registration per CL listener. But Mike has hit on the real zinger -- how does this get automated? I propose a boolean flag -- enable discovery -- per listening CL, then "somehow" the system "automagically" figures out the rest. Notice the hand-waving ... :) ... I dunno how to make it happen, just proposing a "How It Ought To Be" scenario. Jeff On Aug 8, 2007, at 10:53 AM, Michael Demmer wrote: The interfaces are only related to incoming bundles, and so yes, only impact which address the node listens on for connections. You might want to try the bonjour discovery -- If you're on Linux, you can install the avahi-daemon to get the mDNS stack. The ip discovery hasn't been well-tested on multi-homed systems (as you've seen) so it may not set up the bindings properly. No guarantees that bonjour will do better, but it's worth a shot. -m On Aug 8, 2007, at 8:40 AM, Timur Alperovich wrote: Thanks. I'll give that a try. What about specifying local_addr in the interface command? Does that only set the listening address? I'm using the ip module right now. From what I read in the help it appears only bluetooth and ip exist. Are there any others which may be better suited for what I'm doing? Timur On 8/8/07, Michael Demmer wrote: > There's a local_addr option in the link configuration which tells it > which interface to use, if you were to set up an explicit link to the > other node. For example, adding this to node B's config: > > link add l-C 10.0.0.C ALWAYSON TCP local_addr=10.0.0.B > > This assumes that 10.0.0.C is C's ip address on the ad-hoc network > and 10.0.0.B is B's address on the ad-hoc network. The local_addr > option tells the node which address to bind to before connecting to > the other node. > > Given that, I'd love it if the discovery mechanism were to figure out > the situation automatically. Which discovery module are you using? > > -m > > On Aug 8, 2007, at 7:49 AM, Timur Alperovich wrote: > > > Ok, I'm still not sure how to make dtn use a particular interface. > > I tried using the interfaces command and specifying the address. I > > also tried specifying the address in the discovery command. In both > > cases all packets are sent out on only one of the interfaces (eth0 > > in my case). I'd like to also use the wireless and have the node > > act as a bridge on two different subnets. What else do I need to do > > to do that? > > > > Timur > > > > >Hi, > > > > > >I'm trying to use DTN to do some multi-hop networking with the > > dtlsr router and I'm running into a problem with what I >think may > > be the config. I have three nodes. Node A just has an ethernet > > connection. Node B is connected to A over >ethernet and also has a > > wireless card in ad-hoc mode using which it connects to node C. I > > cannot seem to detect any >discoveries over the wireless from node > > B to C. Is there anything I need to add to have dtn work across > > multiple >interfaces? The wireless network is also on a different > > subnet. Would that make a difference? > > > > > >Thank you, > > >Timur > > _______________________________________________ > dtn-users mailing list > dtn-users@mailman.dtnrg.org > http://mailman.dtnrg.org/mailman/listinfo/dtn-users > Received: from gateway0.EECS.Berkeley.EDU (gateway0.EECS.Berkeley.EDU [169.229.60.93]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l78ErOC03399 for ; Wed, 8 Aug 2007 07:53:24 -0700 Received: from [192.168.1.2] (dsl081-061-178.sfo1.dsl.speakeasy.net [64.81.61.178]) (authenticated bits=0) by gateway0.EECS.Berkeley.EDU (8.14.1/8.13.5) with ESMTP id l78FrMeg018937 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Wed, 8 Aug 2007 08:53:24 -0700 (PDT) In-Reply-To: <52950c3f0708080840uf07fc3fg2bf3842df40ab5e4@mail.gmail.com> References: <52950c3f0708071606w5ec0a4efr2a93bcb6e6d8f4f1@mail.gmail.com> <52950c3f0708071611q1b293893gc45938c1a3ce6b58@mail.gmail.com> <52950c3f0708071640g756bdd49l909718ab4cb19f29@mail.gmail.com> <52950c3f0708080749j264f2f95ke37255abf03788c3@mail.gmail.com> <52950c3f0708080840uf07fc3fg2bf3842df40ab5e4@mail.gmail.com> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: multipart/alternative; boundary=Apple-Mail-1--24712719 Message-Id: From: Michael Demmer Subject: Re: [dtn-users] Fwd: dtn on a three node multi-hop X-Applemailsentby: demmer Date: Wed, 8 Aug 2007 08:53:21 -0700 To: dtn-users@mailman.dtnrg.org X-Mailer: Apple Mail (2.752.3) Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: --Apple-Mail-1--24712719 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed The interfaces are only related to incoming bundles, and so yes, only impact which address the node listens on for connections. You might want to try the bonjour discovery -- If you're on Linux, you can install the avahi-daemon to get the mDNS stack. The ip discovery hasn't been well-tested on multi-homed systems (as you've seen) so it may not set up the bindings properly. No guarantees that bonjour will do better, but it's worth a shot. -m On Aug 8, 2007, at 8:40 AM, Timur Alperovich wrote: > Thanks. I'll give that a try. What about specifying local_addr in > the interface command? Does that only set the listening address? > > I'm using the ip module right now. From what I read in the help it > appears only bluetooth and ip exist. Are there any others which may > be better suited for what I'm doing? > > Timur > > On 8/8/07, Michael Demmer wrote: > There's a local_addr option in the link configuration which tells it > which interface to use, if you were to set up an explicit link to the > other node. For example, adding this to node B's config: > > link add l-C 10.0.0.C ALWAYSON TCP local_addr=10.0.0.B > > This assumes that 10.0.0.C is C's ip address on the ad-hoc network > and 10.0.0.B is B's address on the ad-hoc network. The local_addr > option tells the node which address to bind to before connecting to > the other node. > > Given that, I'd love it if the discovery mechanism were to figure out > the situation automatically. Which discovery module are you using? > > -m > > On Aug 8, 2007, at 7:49 AM, Timur Alperovich wrote: > > > Ok, I'm still not sure how to make dtn use a particular interface. > > I tried using the interfaces command and specifying the address. I > > also tried specifying the address in the discovery command. In both > > cases all packets are sent out on only one of the interfaces (eth0 > > in my case). I'd like to also use the wireless and have the node > > act as a bridge on two different subnets. What else do I need to do > > to do that? > > > > Timur > > > > >Hi, > > > > > >I'm trying to use DTN to do some multi-hop networking with the > > dtlsr router and I'm running into a problem with what I >think may > > be the config. I have three nodes. Node A just has an ethernet > > connection. Node B is connected to A over >ethernet and also has a > > wireless card in ad-hoc mode using which it connects to node C. I > > cannot seem to detect any >discoveries over the wireless from node > > B to C. Is there anything I need to add to have dtn work across > > multiple >interfaces? The wireless network is also on a different > > subnet. Would that make a difference? > > > > > >Thank you, > > >Timur > > _______________________________________________ > dtn-users mailing list > dtn-users@mailman.dtnrg.org > http://mailman.dtnrg.org/mailman/listinfo/dtn-users > --Apple-Mail-1--24712719 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=US-ASCII

The interfaces are only = related to incoming bundles, and so yes, only impact which address the = node listens on for connections.

You might want to try the = bonjour discovery -- If you're on Linux, you can install the = avahi-daemon to get the mDNS stack. The ip discovery hasn't been = well-tested on multi-homed systems (as you've seen) so it may not set up = the bindings properly. No guarantees that bonjour will do better, but = it's worth a shot.

-m

On= Aug 8, 2007, at 8:40 AM, Timur Alperovich wrote:

Thanks. = I'll give that a try. What about specifying local_addr in the interface = command? Does that only set the listening address?

I'm using the = ip module right now. =46rom what I read in the help it appears only = bluetooth and ip exist. Are there any others which may be better suited = for what I'm doing?

Timur

On 8/8/07, Michael = Demmer <demmer@cs.berkeley.edu> = wrote:
There's a local_addr option in the link configuration which tells = it
which interface to use, if you were to set up an explicit link to = the
other node. For example, adding this to node B's = config:

link add l-C 10.0.0.C ALWAYSON TCP = local_addr=3D10.0.0.B

This assumes that 10.0.0.C is C's ip = address on the ad-hoc network
and 10.0.0.B is B's address on the = ad-hoc network. The local_addr
option tells the node which address to = bind to before connecting to
the other node.

Given that, I'd = love it if the discovery mechanism were to figure out
the situation = automatically. Which discovery module are you using?

-m

On = Aug 8, 2007, at 7:49 AM, Timur Alperovich wrote:

> Ok, I'm = still not sure how to make dtn use a particular interface.
> I = tried using the interfaces command and specifying the address. I
> = also tried specifying the address in the discovery command. In both =
> cases all packets are sent out on only one of the interfaces = (eth0
> in my case). I'd like to also use the wireless and have = the node
> act as a bridge on two different subnets. What else do = I need to do
> to do that?
>
> Timur
>
> = >Hi,
> >
> >I'm trying to use DTN to do some = multi-hop networking with the
> dtlsr router and I'm running into = a problem with what I >think may
> be the config. I have three = nodes. Node A just has an ethernet
> connection. Node B is = connected to A over >ethernet and also has a
> wireless card in = ad-hoc mode using which it connects to node C. I
> cannot seem to = detect any >discoveries over the wireless from node
> B to C. = Is there anything I need to add to have dtn work across
> multiple = >interfaces? The wireless network is also on a different
> = subnet. Would that make a difference?
> >
> >Thank = you,
> = >Timur

_______________________________________________
dtn-us= ers mailing list
= dtn-users@mailman.dtnrg.org
http://mailma= n.dtnrg.org/mailman/listinfo/dtn-users


= --Apple-Mail-1--24712719-- Received: from rv-out-0910.google.com (rv-out-0910.google.com [209.85.198.187]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l78EeSC03312 for ; Wed, 8 Aug 2007 07:40:28 -0700 Received: by rv-out-0910.google.com with SMTP id f1so148987rvb for ; Wed, 08 Aug 2007 08:40:28 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=LUEeeYQo1pzegWR1+fRtvxA1BzIG8PnPJgn0l8GFf3UwPn7SpjU5NOOYHBcFY4WSGQiKZTRdG4rWmL0bhxbjDgknimboAiy/m1G+V+0AqTGY1BfMKH7O7ndkri0uVc5ZqWcaSpJSouDjiKl4CH/W2LUaTQm5g0W4D3vaCPH8Pbg= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=bY4Kmp88FZIb8+N6X6TCwW9G9eBMlalaSVs1Az/m4vKgAnF730ANPpGOVkhe3GEQowHu2Alsnkz0wfmdyrXk9ofsVrgY1Bn/4QezHaFSH8SOGqApu2zaKIwoIq0wcilYKIDn2pfuyNw5NU5U2kCJrgQBf+1RtlddOWL1CuyB7AU= Received: by 10.141.27.18 with SMTP id e18mr513802rvj.1186587628219; Wed, 08 Aug 2007 08:40:28 -0700 (PDT) Received: by 10.141.51.12 with HTTP; Wed, 8 Aug 2007 08:40:28 -0700 (PDT) Message-ID: <52950c3f0708080840uf07fc3fg2bf3842df40ab5e4@mail.gmail.com> Date: Wed, 8 Aug 2007 11:40:28 -0400 From: "Timur Alperovich" To: dtn-users@mailman.dtnrg.org Subject: Re: [dtn-users] Fwd: dtn on a three node multi-hop In-Reply-To: MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_701_29221967.1186587628209" References: <52950c3f0708071606w5ec0a4efr2a93bcb6e6d8f4f1@mail.gmail.com> <52950c3f0708071611q1b293893gc45938c1a3ce6b58@mail.gmail.com> <52950c3f0708071640g756bdd49l909718ab4cb19f29@mail.gmail.com> <52950c3f0708080749j264f2f95ke37255abf03788c3@mail.gmail.com> Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: ------=_Part_701_29221967.1186587628209 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Thanks. I'll give that a try. What about specifying local_addr in the interface command? Does that only set the listening address? I'm using the ip module right now. From what I read in the help it appears only bluetooth and ip exist. Are there any others which may be better suited for what I'm doing? Timur On 8/8/07, Michael Demmer wrote: > > There's a local_addr option in the link configuration which tells it > which interface to use, if you were to set up an explicit link to the > other node. For example, adding this to node B's config: > > link add l-C 10.0.0.C ALWAYSON TCP local_addr=10.0.0.B > > This assumes that 10.0.0.C is C's ip address on the ad-hoc network > and 10.0.0.B is B's address on the ad-hoc network. The local_addr > option tells the node which address to bind to before connecting to > the other node. > > Given that, I'd love it if the discovery mechanism were to figure out > the situation automatically. Which discovery module are you using? > > -m > > On Aug 8, 2007, at 7:49 AM, Timur Alperovich wrote: > > > Ok, I'm still not sure how to make dtn use a particular interface. > > I tried using the interfaces command and specifying the address. I > > also tried specifying the address in the discovery command. In both > > cases all packets are sent out on only one of the interfaces (eth0 > > in my case). I'd like to also use the wireless and have the node > > act as a bridge on two different subnets. What else do I need to do > > to do that? > > > > Timur > > > > >Hi, > > > > > >I'm trying to use DTN to do some multi-hop networking with the > > dtlsr router and I'm running into a problem with what I >think may > > be the config. I have three nodes. Node A just has an ethernet > > connection. Node B is connected to A over >ethernet and also has a > > wireless card in ad-hoc mode using which it connects to node C. I > > cannot seem to detect any >discoveries over the wireless from node > > B to C. Is there anything I need to add to have dtn work across > > multiple >interfaces? The wireless network is also on a different > > subnet. Would that make a difference? > > > > > >Thank you, > > >Timur > > _______________________________________________ > dtn-users mailing list > dtn-users@mailman.dtnrg.org > http://mailman.dtnrg.org/mailman/listinfo/dtn-users > ------=_Part_701_29221967.1186587628209 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Thanks. I'll give that a try. What about specifying local_addr in the interface command? Does that only set the listening address?

I'm using the ip module right now. From what I read in the help it appears only bluetooth and ip exist. Are there any others which may be better suited for what I'm doing?

Timur

On 8/8/07, Michael Demmer <demmer@cs.berkeley.edu> wrote:
There's a local_addr option in the link configuration which tells it
which interface to use, if you were to set up an explicit link to the
other node. For example, adding this to node B's config:

link add l-C 10.0.0.C ALWAYSON TCP local_addr=10.0.0.B

This assumes that 10.0.0.C is C's ip address on the ad-hoc network
and 10.0.0.B is B's address on the ad-hoc network. The local_addr
option tells the node which address to bind to before connecting to
the other node.

Given that, I'd love it if the discovery mechanism were to figure out
the situation automatically. Which discovery module are you using?

-m

On Aug 8, 2007, at 7:49 AM, Timur Alperovich wrote:

> Ok, I'm still not sure how to make dtn use a particular interface.
> I tried using the interfaces command and specifying the address. I
> also tried specifying the address in the discovery command. In both
> cases all packets are sent out on only one of the interfaces (eth0
> in my case). I'd like to also use the wireless and have the node
> act as a bridge on two different subnets. What else do I need to do
> to do that?
>
> Timur
>
> >Hi,
> >
> >I'm trying to use DTN to do some multi-hop networking with the
> dtlsr router and I'm running into a problem with what I >think may
> be the config. I have three nodes. Node A just has an ethernet
> connection. Node B is connected to A over >ethernet and also has a
> wireless card in ad-hoc mode using which it connects to node C. I
> cannot seem to detect any >discoveries over the wireless from node
> B to C. Is there anything I need to add to have dtn work across
> multiple >interfaces? The wireless network is also on a different
> subnet. Would that make a difference?
> >
> >Thank you,
> >Timur

_______________________________________________
dtn-users mailing list
dtn-users@mailman.dtnrg.org
http://mailman.dtnrg.org/mailman/listinfo/dtn-users

------=_Part_701_29221967.1186587628209-- Received: from gateway0.EECS.Berkeley.EDU (gateway0.EECS.Berkeley.EDU [169.229.60.93]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l78E2xC03044 for ; Wed, 8 Aug 2007 07:02:59 -0700 Received: from [192.168.1.2] (dsl081-061-178.sfo1.dsl.speakeasy.net [64.81.61.178]) (authenticated bits=0) by gateway0.EECS.Berkeley.EDU (8.14.1/8.13.5) with ESMTP id l78F2v1r018237 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Wed, 8 Aug 2007 08:02:59 -0700 (PDT) In-Reply-To: <52950c3f0708080749j264f2f95ke37255abf03788c3@mail.gmail.com> References: <52950c3f0708071606w5ec0a4efr2a93bcb6e6d8f4f1@mail.gmail.com> <52950c3f0708071611q1b293893gc45938c1a3ce6b58@mail.gmail.com> <52950c3f0708071640g756bdd49l909718ab4cb19f29@mail.gmail.com> <52950c3f0708080749j264f2f95ke37255abf03788c3@mail.gmail.com> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: Content-Transfer-Encoding: 7bit From: Michael Demmer Subject: Re: [dtn-users] Fwd: dtn on a three node multi-hop X-Applemailsentby: demmer Date: Wed, 8 Aug 2007 08:02:56 -0700 To: dtn-users@mailman.dtnrg.org X-Mailer: Apple Mail (2.752.3) Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: There's a local_addr option in the link configuration which tells it which interface to use, if you were to set up an explicit link to the other node. For example, adding this to node B's config: link add l-C 10.0.0.C ALWAYSON TCP local_addr=10.0.0.B This assumes that 10.0.0.C is C's ip address on the ad-hoc network and 10.0.0.B is B's address on the ad-hoc network. The local_addr option tells the node which address to bind to before connecting to the other node. Given that, I'd love it if the discovery mechanism were to figure out the situation automatically. Which discovery module are you using? -m On Aug 8, 2007, at 7:49 AM, Timur Alperovich wrote: > Ok, I'm still not sure how to make dtn use a particular interface. > I tried using the interfaces command and specifying the address. I > also tried specifying the address in the discovery command. In both > cases all packets are sent out on only one of the interfaces (eth0 > in my case). I'd like to also use the wireless and have the node > act as a bridge on two different subnets. What else do I need to do > to do that? > > Timur > > >Hi, > > > >I'm trying to use DTN to do some multi-hop networking with the > dtlsr router and I'm running into a problem with what I >think may > be the config. I have three nodes. Node A just has an ethernet > connection. Node B is connected to A over >ethernet and also has a > wireless card in ad-hoc mode using which it connects to node C. I > cannot seem to detect any >discoveries over the wireless from node > B to C. Is there anything I need to add to have dtn work across > multiple >interfaces? The wireless network is also on a different > subnet. Would that make a difference? > > > >Thank you, > >Timur Received: from rv-out-0910.google.com (rv-out-0910.google.com [209.85.198.190]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l78DnYC02926 for ; Wed, 8 Aug 2007 06:49:34 -0700 Received: by rv-out-0910.google.com with SMTP id f1so137942rvb for ; Wed, 08 Aug 2007 07:49:29 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=LrkJSV9pj5Q/5c5c1kdxbysmv19Xxy0Y4HZjsDe22epYvLydldwc+2hSidNjirTO9J/eWL1wrvfInggge5cwZ8VmsVccd9Vn9de5DK//9gomEmLRFs9Kh2QDcGgA1ePWEoID6s0oXn2MLeSeR0HbP88GR8FTqB4uLZlYcmGZ1iY= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=TbU4hE36ZNqIZQEzivHemuMNLkcCywn8q0xu+XAYn05V8D8R3qYNAjbl5HDN7E2VunDtcpWEiyDEODT7nkRo9HS0wGowdnB/rE1f43esW0CHnGZAM6mwF2rx6R0IvyLqIZMFjsfzQuMry5yEJ6osqqwLvfTrZzvsMaFXYOuAjGs= Received: by 10.141.53.15 with SMTP id f15mr172420rvk.1186584568949; Wed, 08 Aug 2007 07:49:28 -0700 (PDT) Received: by 10.141.51.12 with HTTP; Wed, 8 Aug 2007 07:49:28 -0700 (PDT) Message-ID: <52950c3f0708080749j264f2f95ke37255abf03788c3@mail.gmail.com> Date: Wed, 8 Aug 2007 10:49:28 -0400 From: "Timur Alperovich" To: dtn-users@mailman.dtnrg.org In-Reply-To: <52950c3f0708071640g756bdd49l909718ab4cb19f29@mail.gmail.com> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_517_11235481.1186584568931" References: <52950c3f0708071606w5ec0a4efr2a93bcb6e6d8f4f1@mail.gmail.com> <52950c3f0708071611q1b293893gc45938c1a3ce6b58@mail.gmail.com> <52950c3f0708071640g756bdd49l909718ab4cb19f29@mail.gmail.com> Subject: [dtn-users] Fwd: dtn on a three node multi-hop Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: ------=_Part_517_11235481.1186584568931 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Ok, I'm still not sure how to make dtn use a particular interface. I tried using the interfaces command and specifying the address. I also tried specifying the address in the discovery command. In both cases all packets are sent out on only one of the interfaces (eth0 in my case). I'd like to also use the wireless and have the node act as a bridge on two different subnets. What else do I need to do to do that? Timur >Hi, > >I'm trying to use DTN to do some multi-hop networking with the dtlsr router and I'm running into a problem with what I >think may be the config. I have three nodes. Node A just has an ethernet connection. Node B is connected to A over >ethernet and also has a wireless card in ad-hoc mode using which it connects to node C. I cannot seem to detect any >discoveries over the wireless from node B to C. Is there anything I need to add to have dtn work across multiple >interfaces? The wireless network is also on a different subnet. Would that make a difference? > >Thank you, >Timur ------=_Part_517_11235481.1186584568931 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Ok, I'm still not sure how to make dtn use a particular interface. I tried using the interfaces command and specifying the address. I also tried specifying the address in the discovery command. In both cases all packets are sent out on only one of the interfaces (eth0 in my case). I'd like to also use the wireless and have the node act as a bridge on two different subnets. What else do I need to do to do that?

Timur

>Hi,
>
>I'm trying to use DTN to do some multi-hop networking with the dtlsr router and I'm running into a problem with what I >think may be the config. I have three nodes. Node A just has an ethernet connection. Node B is connected to A over >ethernet and also has a wireless card in ad-hoc mode using which it connects to node C. I cannot seem to detect any >discoveries over the wireless from node B to C. Is there anything I need to add to have dtn work across multiple >interfaces? The wireless network is also on a different subnet. Would that make a difference?
>
>Thank you,
>Timur
------=_Part_517_11235481.1186584568931-- Received: from gateway0.EECS.Berkeley.EDU (gateway0.EECS.Berkeley.EDU [169.229.60.93]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l77NhjC27963; Tue, 7 Aug 2007 16:43:45 -0700 Received: from [10.212.2.161] (bldmz-nat-161-182.berkeley.intel-research.net [12.155.161.182]) (authenticated bits=0) by gateway0.EECS.Berkeley.EDU (8.14.1/8.13.5) with ESMTP id l780hiIf014722 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Tue, 7 Aug 2007 17:43:45 -0700 (PDT) Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: Content-Transfer-Encoding: 7bit From: Michael Demmer X-Applemailsentby: demmer Date: Tue, 7 Aug 2007 17:43:41 -0700 To: dtn-users@mailman.dtnrg.org, dtn-dev X-Mailer: Apple Mail (2.752.3) Subject: [dtn-users] changing version control systems Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: I'm considering changing the main DTN reference implementation source repository from CVS to subversion[1] or mercurial[2]. There are numerous reasons why these other systems are better version control mechanisms than CVS, including the ability to rename files, grouping modifications into logical changesets, directories that are editable in the VC system, and better support for branching, to name a few. Mercurial adds the additional advantages of being a distributed rather than centralized system. On the other hand, switching will likely entail transition pain, and some people may not be familiar with the new systems. In particular, mercurial has a very different usage model which will change how people interact with the source repository. The intent of this mail is to get a sense of the overall temperature of the DTN user base out there with respect to this change: Do you think it's a great idea? Is it a terrible idea? Will it impede your interactions with the code? Do you have horror stories about subversion and/or mercurial? Does anyone have a strong preference for one versus the other? -m --- [1] http://subversion.tigris.org/ [2] http://www.selenic.com/mercurial/wiki/ Received: from rv-out-0910.google.com (rv-out-0910.google.com [209.85.198.191]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l77Mf0C27610 for ; Tue, 7 Aug 2007 15:41:00 -0700 Received: by rv-out-0910.google.com with SMTP id k20so1386896rvb for ; Tue, 07 Aug 2007 16:40:57 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=Ct5h6drR34zJFkgQaop9mDWj6Tc1ycT16qsuztUzL+058hyHd+91Q3ZB5CT3hhyKxECDjE3Q8q7HbE7lXb8DPeudECYDkgwnmusMBUyI6RIywaCYD6dteaf0siXUmQNCXjRaondQKicMke04T89yiP4x0hZ6Vsc+5elTsKvM/3o= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=Gr2RbRuZ30QNJGouTfpkc5LaOwN/4k8gF+R7mzq6YeFD+0hBGi+u2UxqB1IaxFYr6dVqrBUw4inrZaBKRKtTfTIDf1ETYpdd824FkemFc0x8YmmopTIwd+wt3cQDjKzSmSIyOyWyp4zXPKhV0ynYLj+vKy5nMMKOh/jSz0wE6uw= Received: by 10.141.172.6 with SMTP id z6mr2754098rvo.1186530057620; Tue, 07 Aug 2007 16:40:57 -0700 (PDT) Received: by 10.141.51.12 with HTTP; Tue, 7 Aug 2007 16:40:57 -0700 (PDT) Message-ID: <52950c3f0708071640g756bdd49l909718ab4cb19f29@mail.gmail.com> Date: Tue, 7 Aug 2007 19:40:57 -0400 From: "Timur Alperovich" To: dtn-users@mailman.dtnrg.org In-Reply-To: <52950c3f0708071611q1b293893gc45938c1a3ce6b58@mail.gmail.com> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_34006_17606343.1186530057591" References: <52950c3f0708071606w5ec0a4efr2a93bcb6e6d8f4f1@mail.gmail.com> <52950c3f0708071611q1b293893gc45938c1a3ce6b58@mail.gmail.com> Subject: [dtn-users] Fwd: dtn on a three node multi-hop Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: ------=_Part_34006_17606343.1186530057591 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Never mind. I realized that I have to specify addresses for the interfaces. Timur >Hi, > >I'm trying to use DTN to do some multi-hop networking with the dtlsr router and I'm running into a problem with what I >think may be the config. I have three nodes. Node A just has an ethernet connection. Node B is connected to A over >ethernet and also has a wireless card in ad-hoc mode using which it connects to node C. I cannot seem to detect any >discoveries over the wireless from node B to C. Is there anything I need to add to have dtn work across multiple >interfaces? The wireless network is also on a different subnet. Would that make a difference? > >Thank you, >Timur ------=_Part_34006_17606343.1186530057591 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Never mind. I realized that I have to specify addresses for the interfaces.

Timur

>Hi,
>
>I'm trying to use DTN to do some multi-hop networking with the dtlsr router and I'm running into a problem with what I >think may be the config. I have three nodes. Node A just has an ethernet connection. Node B is connected to A over >ethernet and also has a wireless card in ad-hoc mode using which it connects to node C. I cannot seem to detect any >discoveries over the wireless from node B to C. Is there anything I need to add to have dtn work across multiple >interfaces? The wireless network is also on a different subnet. Would that make a difference?
>
>Thank you,
>Timur
------=_Part_34006_17606343.1186530057591-- Received: from rv-out-0910.google.com (rv-out-0910.google.com [209.85.198.190]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l77MBPC27436 for ; Tue, 7 Aug 2007 15:11:25 -0700 Received: by rv-out-0910.google.com with SMTP id k20so1380900rvb for ; Tue, 07 Aug 2007 16:11:25 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=cCZXxUzls7wdxCnP/8Zd5S+86vQ/7okOB9MsjG4YFPk58Q0YjfPS6SZiMlpcSTojRzYOhfumThRF5yuaVbijyS8rXZBnb4vLWaMQ+ZxqUAXfZTHt+RqT9Bn++koLi/nvZlY1VQNQL0ecgpDz7DdpbtJC6lY0M63tEdwxpa/K/d8= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=Bu3S75coLcySFCCd92QRFIyXMU+jwLKEawT2TumF1hRQp9hAYuli+CYY7PqOT3lhocAYi/xG++ZKAkAtmiMZ1DcKWt0lLHL292OhaI/uB7/boqhESJIVPL0gR6d6Yd8mDZpy50stp+Rx6aD3764ntX1J12DHZoPBQMyS7so5vXw= Received: by 10.141.27.18 with SMTP id e18mr2747541rvj.1186528285100; Tue, 07 Aug 2007 16:11:25 -0700 (PDT) Received: by 10.141.51.12 with HTTP; Tue, 7 Aug 2007 16:11:25 -0700 (PDT) Message-ID: <52950c3f0708071611q1b293893gc45938c1a3ce6b58@mail.gmail.com> Date: Tue, 7 Aug 2007 19:11:25 -0400 From: "Timur Alperovich" To: dtn-users@mailman.dtnrg.org In-Reply-To: <52950c3f0708071606w5ec0a4efr2a93bcb6e6d8f4f1@mail.gmail.com> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_33962_27565197.1186528285077" References: <52950c3f0708071606w5ec0a4efr2a93bcb6e6d8f4f1@mail.gmail.com> Subject: [dtn-users] dtn on a three node multi-hop Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: ------=_Part_33962_27565197.1186528285077 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Hi, I'm trying to use DTN to do some multi-hop networking with the dtlsr router and I'm running into a problem with what I think may be the config. I have three nodes. Node A just has an ethernet connection. Node B is connected to A over ethernet and also has a wireless card in ad-hoc mode using which it connects to node C. I cannot seem to detect any discoveries over the wireless from node B to C. Is there anything I need to add to have dtn work across multiple interfaces? The wireless network is also on a different subnet. Would that make a difference? Thank you, Timur ------=_Part_33962_27565197.1186528285077 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Hi,

I'm trying to use DTN to do some multi-hop networking with the dtlsr router and I'm running into a problem with what I think may be the config. I have three nodes. Node A just has an ethernet connection. Node B is connected to A over ethernet and also has a wireless card in ad-hoc mode using which it connects to node C. I cannot seem to detect any discoveries over the wireless from node B to C. Is there anything I need to add to have dtn work across multiple interfaces? The wireless network is also on a different subnet. Would that make a difference?

Thank you,
Timur
 ------=_Part_33962_27565197.1186528285077-- Received: from gateway0.EECS.Berkeley.EDU (gateway0.EECS.Berkeley.EDU [169.229.60.93]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l77Hl6C25555 for ; Tue, 7 Aug 2007 10:47:06 -0700 Received: from [10.212.2.161] (bldmz-nat-161-182.berkeley.intel-research.net [12.155.161.182]) (authenticated bits=0) by gateway0.EECS.Berkeley.EDU (8.14.1/8.13.5) with ESMTP id l77Il4w3009253 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Tue, 7 Aug 2007 11:47:05 -0700 (PDT) In-Reply-To: <1186075310.46b212aed4cef@webmail.ohio.edu> References: <1186075310.46b212aed4cef@webmail.ohio.edu> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: <2E1BE7C2-FBEA-4655-9DB6-DBF634B2E431@cs.berkeley.edu> Cc: dtn-users@mailman.dtnrg.org Content-Transfer-Encoding: 7bit From: Michael Demmer X-Applemailsentby: demmer Date: Tue, 7 Aug 2007 11:47:00 -0700 To: jw280601@ohio.edu X-Mailer: Apple Mail (2.752.3) Subject: [dtn-users] Re: DTN2 Compilation Problem Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: I just tried this test, and to me it looks like the configure script propagated the --with-tcl argument to oasys properly. Can you send me the output from your failed configure invocation and the contents of the generated config.log so I can help see what's going on? Also, make sure you passed an absolute path to --with-tcl if you want it to work properly. -m On Aug 2, 2007, at 10:21 AM, jw280601@ohio.edu wrote: > There's a problem when compiling the DTN2 code. If you specify a > different > Tcl library location for compiling DTN2, it won't propagate the > change to > Oasys (which will them fail as it can't find the library). I > haven't tested > to see if this is fixed by the warning to cache things found in ./ > configure (I > suspect it is), but a note somewhere might be useful. > > Doing a ./configure on Oasys with the proper location seemed to fix > the > problem. > > Jim Wyllie Received: from rv-out-0910.google.com (rv-out-0910.google.com [209.85.198.186]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l72Ii5C25954 for ; Thu, 2 Aug 2007 11:44:05 -0700 Received: by rv-out-0910.google.com with SMTP id f1so440998rvb for ; Thu, 02 Aug 2007 12:44:05 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=CL4HRd9yNw8MN9h89IX/cdbUbF5UEmEuzU+ICcitor5ftxvQOU1NmZOOHueG9bbsgXsq9IDlVH7+FOn6gMoaLUTbReSS2R4az8n987h+aP5MnXaXOVxuQEW/OI1f0ikROJzUBMmvgDkjI4jAIEO8b3dUOT3e449t2jzOGUmG+3c= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=ELajSmPDavmR+oSN/tJB9g65mtd1ZiuCzH2THuV173oWKSL48QkySj247ye2IlALPfJMNUmPpeqIoVZ4qUH8O1zngKilXN6JGJcP/peg3Aeegm0hvRWmxKn/0RZ8Wp1J5qMDc/K8PWHlam8obJulDzJOtvTzOGv3ocSK2RUMFK0= Received: by 10.141.28.12 with SMTP id f12mr782729rvj.1186083844810; Thu, 02 Aug 2007 12:44:04 -0700 (PDT) Received: by 10.141.51.12 with HTTP; Thu, 2 Aug 2007 12:44:04 -0700 (PDT) Message-ID: <52950c3f0708021244u1c3d6045w90db3f6cfd68407c@mail.gmail.com> Date: Thu, 2 Aug 2007 15:44:04 -0400 From: "Timur Alperovich" To: dtn-users@mailman.dtnrg.org Subject: Re: [dtn-users] tcp neighbor discovery In-Reply-To: <52950c3f0708020936k6dbb760am6cf3961706839e12@mail.gmail.com> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_19175_10627104.1186083844800" References: <52950c3f0707301019i2800b93ft2af045e9c2c56a94@mail.gmail.com> <4BBA0AB4-8EC5-4255-B420-7352E4700C6D@cs.berkeley.edu> <52950c3f0707301157h430d89f5h83500a122cb30cf5@mail.gmail.com> <1D085F75-B987-4E5F-AF9B-ECD35D957B38@cs.berkeley.edu> <52950c3f0708020936k6dbb760am6cf3961706839e12@mail.gmail.com> Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: ------=_Part_19175_10627104.1186083844800 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Never mind actually. I realized that I needed to configure discovery and it's working now. Thanks for all the tips. Timur On 8/2/07, Timur Alperovich wrote: > > Mike, > > Thanks for the help. I did not see dtlsr listed in the list of routers in > the config file and that's why I was a little confused and tried to use > linkstate. I ran into another issue now, which I hope is not a big deal > either. I tried using the dtlsr router by simply setting the route type to > dtlsr, but I'm not sure what else, if anything, I need to add for nodes to > discover their neighbors. Is there any chance you could send me a sample > config file? > > Thank you, > Timur > > On 7/30/07, Michael Demmer wrote: > > > > Oh... you want the "dtlsr" router, not the "linkstate" router. That > > latter one is very old (and should probably just be dropped from the > > repository). > > > > -mike > > > ------=_Part_19175_10627104.1186083844800 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Never mind actually. I realized that I needed to configure discovery and it's working now. Thanks for all the tips.

Timur

On 8/2/07, Timur Alperovich <timur.alperovich@gmail.com> wrote:
Mike,

Thanks for the help. I did not see dtlsr listed in the list of routers in the config file and that's why I was a little confused and tried to use linkstate. I ran into another issue now, which I hope is not a big deal either. I tried using the dtlsr router by simply setting the route type to dtlsr, but I'm not sure what else, if anything, I need to add for nodes to discover their neighbors. Is there any chance you could send me a sample config file?

Thank you,
Timur

On 7/30/07, Michael Demmer < demmer@cs.berkeley.edu> wrote:
Oh... you want the "dtlsr" router, not the "linkstate" router. That latter one is very old (and should probably just be dropped from the repository).

-mike

------=_Part_19175_10627104.1186083844800-- Received: from rv-out-0910.google.com (rv-out-0910.google.com [209.85.198.189]) by webbie.berkeley.intel-research.net (8.11.6/8.11.6) with ESMTP id l72FawC24754 for ; Thu, 2 Aug 2007 08:36:58 -0700 Received: by rv-out-0910.google.com with SMTP id f1so407833rvb for ; Thu, 02 Aug 2007 09:36:55 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=Q29CESrcdHHIL7ql1Nz0YyHhMGQ5JT8KGWpFLp+G52dH1UAKMIx8ocOxdD3x08giztJXhwcLwWdGOtux65tCjy7yQqMqZaz8vzgCrsNwQM47mpbPvzsaTRrdy/AupESlGtVhS73VXgfbOON6YXESjjdvTzSLifq/Q8denF6Cau8= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=fY/7GgDuAZtTkezSbcHUI5YFtDSBhWCwGXpnCXGgKkZRbjqYZabGLd3KmxN0zSpjMWuR1Ft6zhEkdFplTF2ILefPclbjkB3rAQQBQiWC/HGX0LogTU5Ixd7t0zaoXQJ8TXDfHS5EKs/f/WuTiZ15UmrX0tTaQs+nCf8eHlAiAWI= Received: by 10.141.197.18 with SMTP id z18mr715894rvp.1186072614368; Thu, 02 Aug 2007 09:36:54 -0700 (PDT) Received: by 10.141.51.12 with HTTP; Thu, 2 Aug 2007 09:36:54 -0700 (PDT) Message-ID: <52950c3f0708020936k6dbb760am6cf3961706839e12@mail.gmail.com> Date: Thu, 2 Aug 2007 12:36:54 -0400 From: "Timur Alperovich" To: dtn-users@mailman.dtnrg.org Subject: Re: [dtn-users] tcp neighbor discovery In-Reply-To: <1D085F75-B987-4E5F-AF9B-ECD35D957B38@cs.berkeley.edu> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_18534_7071407.1186072614355" References: <52950c3f0707301019i2800b93ft2af045e9c2c56a94@mail.gmail.com> <4BBA0AB4-8EC5-4255-B420-7352E4700C6D@cs.berkeley.edu> <52950c3f0707301157h430d89f5h83500a122cb30cf5@mail.gmail.com> <1D085F75-B987-4E5F-AF9B-ECD35D957B38@cs.berkeley.edu> Sender: dtn-users-admin@mailman.dtnrg.org Errors-To: dtn-users-admin@mailman.dtnrg.org X-BeenThere: dtn-users@mailman.dtnrg.org X-Mailman-Version: 2.0.13 Precedence: bulk Reply-To: dtn-users@mailman.dtnrg.org List-Unsubscribe: , List-Id: discussion/answers list for users of the dtn reference implementation List-Post: List-Help: List-Subscribe: , List-Archive: ------=_Part_18534_7071407.1186072614355 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Mike, Thanks for the help. I did not see dtlsr listed in the list of routers in the config file and that's why I was a little confused and tried to use linkstate. I ran into another issue now, which I hope is not a big deal either. I tried using the dtlsr router by simply setting the route type to dtlsr, but I'm not sure what else, if anything, I need to add for nodes to discover their neighbors. Is there any chance you could send me a sample config file? Thank you, Timur On 7/30/07, Michael Demmer wrote: > > Oh... you want the "dtlsr" router, not the "linkstate" router. That latter > one is very old (and should probably just be dropped from the repository). > > -mike > ------=_Part_18534_7071407.1186072614355 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Mike,

Thanks for the help. I did not see dtlsr listed in the list of routers in the config file and that's why I was a little confused and tried to use linkstate. I ran into another issue now, which I hope is not a big deal either. I tried using the dtlsr router by simply setting the route type to dtlsr, but I'm not sure what else, if anything, I need to add for nodes to discover their neighbors. Is there any chance you could send me a sample config file?

Thank you,
Timur

On 7/30/07, Michael Demmer <demmer@cs.berkeley.edu> wrote:
Oh... you want the "dtlsr" router, not the "linkstate" router. That latter one is very old (and should probably just be dropped from the repository).

-mike
------=_Part_18534_7071407.1186072614355--