From dbri.tcc at gmail.com Tue Mar 11 16:04:42 2008 From: dbri.tcc at gmail.com (Don Bright) Date: Tue, 11 Mar 2008 10:04:42 -0500 Subject: [Udpcast] description of how i used udpcast Message-ID: Hello Just wanted to describe how I used udpcast. I worked in a library, we had about 40 machines of different types. They were all running Windows XP. I muddled through Windows Sysprep and figured out how to make a single image that worked with all of them. Then, I installed ubuntu on every single machine, on a second partition... it was about 15 gigabytes. Within this partition, I had a gigantic file - it was an linux-ntfs image of that Windows XP partition. If I remember, it was between 5 and 10 gigabytes. Whenever I wanted to 'update' all the machines, with windows updates, software installation, tweaks, etc, I would just do it on one windows machine, reboot into ubunutu, image the windows partition down into a big image file within the linux partition. Then I would the reboot all the machines into ubuntu, then broadcast that windows image file, and everything was updated, in like 45 minutes. And If I wanted to wipe out any possible viruses, etc, I just unpacked an old clean image, and broadcast it. I had python scripts that basically automated the entire thing, it took like 10 keystrokes to broadcast the image and like 5 keystrokes to set up a machine to accept an image, and it had menus and everything. Just wanted to put this out there, in case some jerk-off tries to patent this relatively obvious idea. And to shout out to the udpcast people, rock on. -------------- next part -------------- An HTML attachment was scrubbed... URL: From brian at polibyte.com Thu Mar 13 00:05:28 2008 From: brian at polibyte.com (brian at polibyte.com) Date: Wed, 12 Mar 2008 19:05:28 -0400 Subject: [Udpcast] Request - include netcat Message-ID: <20080312190528.tlsucpsnuoocksgk@polibyte.com> Hi, It would be nice if netcat were included in the CD and PXE images. Busybox has a version, so adding it shouldn't take much. All the best, Brian From james.tipler at stcatz.ox.ac.uk Thu Mar 13 16:43:28 2008 From: james.tipler at stcatz.ox.ac.uk (James Tipler) Date: Thu, 13 Mar 2008 15:43:28 -0000 Subject: [Udpcast] newbie questions: Message-ID: <001001c88520$fb74c280$00324381@catz5000> Hi, I apologize if this is a stupid question - actually, I rather hope it is because if I've just missed something somewhere in the documentation then this should be easy to sort out. I've got about 30 work stations here that are used by conferences a dozen or so times a year. Each time they come in they manage to make a mess of them in one way or another, so each workstation gets re-imaged once they've gone away. Currently, we do this with a little linux partition with G4L on it and a little tweaking of the Windows NT boot loader. (Basically, run batch file in windows - machine goes down, linux comes up, script runs, disk image.gz is pulled from FTP server and written to the windows partition, machine goes down, windows comes up, job'sagood'un) This works rather nicely, but it takes a painfully long time as the FTP server struggles under the load of 30 connections. UDP-cast looks perfect for solving this problem, but I can't seem to make it work. This is the command I've got: (from the FTP server - windows 2003 machine) Udp-sender.exe -file c:\bootfiles\image.gz (from the client machines linux command line) Udp-receiver -pipe "gzip -dc" -file /dev/sda1 The network infrastructure is very simple. All the machines involved are in the same subnet 129.67.50.xx (in fact they're all on the same switch to try and keep this as fast as possible) The UDP sender starts to run and awaits connections, but never seems to get any. The UDP receivers sit and wait for a control connection from the server, but never get any either. What am I doing wrong? -------------- next part -------------- An HTML attachment was scrubbed... URL: From mikes at kuentos.guam.net Thu Mar 13 23:49:04 2008 From: mikes at kuentos.guam.net (Michael D. Setzer II) Date: Fri, 14 Mar 2008 08:49:04 +1000 Subject: [Udpcast] newbie questions: In-Reply-To: <001001c88520$fb74c280$00324381@catz5000> References: <001001c88520$fb74c280$00324381@catz5000> Message-ID: <47DA3C00.17254.69DC689@mikes.kuentos.guam.net> On 13 Mar 2008 at 15:43, James Tipler wrote: > > Hi, I apologize if this is a stupid question – actually, I rather hope it is because if I’ve just missed > something somewhere in the documentation then this should be easy to sort out. > > I’ve got about 30 work stations here that are used by conferences a dozen or so times a year. > Each time they come in they manage to make a mess of them in one way or another, so each > workstation gets re-imaged once they’ve gone away. Currently, we do this with a little linux > partition with G4L on it and a little tweaking of the Windows NT boot loader. (Basically, run batch > file in windows – machine goes down, linux comes up, script runs, disk image.gz is pulled from > FTP server and written to the windows partition, machine goes down, windows comes up, > job’sagood’un) > > This works rather nicely, but it takes a painfully long time as the FTP server struggles under the > load of 30 connections. UDP-cast looks perfect for solving this problem, but I can’t seem to make > it work. This is the command I’ve got: > > (from the FTP server –windows 2003 machine) > Udp-sender.exe –file c:\bootfiles\image.gz > > (from the client machines linux command line) > Udp-receiver –pipe “gzip –dc” –file /dev/sda1 > > The network infrastructure is very simple. All the machines involved are in the same subnet > 129.67.50.xx (in fact they’re all on the same switch to try and keep this as fast as possible) > > The UDP sender starts to run and awaits connections, but never seems to get any. > The UDP receivers sit and wait for a control connection from the server, but never get any either. > > What am I doing wrong? > Not a 100% sure? But here are some things. First off, what kind of switch do you have? I had no problems with my old hubs, then MIS replaced them with 3COM switches, and udpcast stopped working. It would work with a single set of machine, but not multiple. Turned out to be an IGMP setting if I recall correctly. Second, what does it say exactly it show on the sender and reciever screens concerning the multicast- G4L does have the option to receive udpcast, and checking the lines, with lzop compression it is `lzop -d -c -` and with gzip it shows `gzip -c - ` but that is with linux, and it might just be the difference in how windows does it. So, it could just be the missing -. Being the mainter of G4L, it make use of udpcast every week. I generally make updates to my lab machines, and create a new g4l backup image, and the use udpcast to image all the other machines in the lab. So, one ftp download to a sinfle machine, and then multi-cast to all the others. I created a custome udpcast send disk, and 7 copies of a udp-recieve disk. Just put the sender disk in the one machine, and it has all options set, and is ready to go. Then boot other machines from the reciever disk, and then hit enter on any machine went all are ready. I setup these CDs to do a whole disk image, so no inputed needed at all. Works just great. Another options. In my lab, the XP partition is always get messed up or changes, and we were recently getting this kavo virus being spread by flahes, and Norton was not stopping it at all. In this case, my XP partition imagge is about 6GB, so I just copied it to the Linux partition, which is 50% of the 80GB drive. So, it had plenty of free space. I then have it do a local ntfsclone restore to the XP partition, and can reimage the windows in about 10 minutes. NTFSCLONE is faster, since it only copies used data, and by not downloadin the image to all machines at once puts no load on the network. So, thre are some things to look at. +----------------------------------------------------------+ Michael D. Setzer II - Computer Science Instructor Guam Community College Computer Center mailto:mikes at kuentos.guam.net mailto:msetzerii at gmail.com http://www.guam.net/home/mikes Guam - Where America's Day Begins +----------------------------------------------------------+ http://setiathome.berkeley.edu (Original) Number of Seti Units Returned: 19,471 Processing time: 32 years, 290 days, 12 hours, 58 minutes (Total Hours: 287,489) BOINC at HOME CREDITS SETI 4,925,706.239266 | EINSTEIN 1,464,634.787568 | ROSETTA 428,541.225917 From james.tipler at stcatz.ox.ac.uk Fri Mar 14 09:43:26 2008 From: james.tipler at stcatz.ox.ac.uk (James Tipler) Date: Fri, 14 Mar 2008 08:43:26 -0000 Subject: [Udpcast] newbie questions: In-Reply-To: <47DA3C00.17254.69DC689@mikes.kuentos.guam.net> References: <001001c88520$fb74c280$00324381@catz5000> <47DA3C00.17254.69DC689@mikes.kuentos.guam.net> Message-ID: <001e01c885af$785a8b70$00324381@catz5000> Thanks for replying so fast! Actually, you're point about the switch is spot on. It's a 3com, and spending a few minutes poking around in it's config last night sorted everything out - should have checked it before posting here really! Thanks again. -----Original Message----- From: Michael D. Setzer II [mailto:mikes at kuentos.guam.net] Sent: 13 March 2008 22:49 To: James Tipler; udpcast at udpcast.linux.lu Subject: Re: [Udpcast] newbie questions: On 13 Mar 2008 at 15:43, James Tipler wrote: > > Hi, I apologize if this is a stupid question - actually, I rather hope it is because if I've just missed > something somewhere in the documentation then this should be easy to sort out. > > I've got about 30 work stations here that are used by conferences a dozen or so times a year. > Each time they come in they manage to make a mess of them in one way or another, so each > workstation gets re-imaged once they've gone away. Currently, we do this with a little linux > partition with G4L on it and a little tweaking of the Windows NT boot loader. (Basically, run batch > file in windows - machine goes down, linux comes up, script runs, disk image.gz is pulled from > FTP server and written to the windows partition, machine goes down, windows comes up, > job'sagood'un) > > This works rather nicely, but it takes a painfully long time as the FTP server struggles under the > load of 30 connections. UDP-cast looks perfect for solving this problem, but I can't seem to make > it work. This is the command I've got: > > (from the FTP server -windows 2003 machine) > Udp-sender.exe -file c:\bootfiles\image.gz > > (from the client machines linux command line) > Udp-receiver -pipe "gzip -dc" -file /dev/sda1 > > The network infrastructure is very simple. All the machines involved are in the same subnet > 129.67.50.xx (in fact they're all on the same switch to try and keep this as fast as possible) > > The UDP sender starts to run and awaits connections, but never seems to get any. > The UDP receivers sit and wait for a control connection from the server, but never get any either. > > What am I doing wrong? > Not a 100% sure? But here are some things. First off, what kind of switch do you have? I had no problems with my old hubs, then MIS replaced them with 3COM switches, and udpcast stopped working. It would work with a single set of machine, but not multiple. Turned out to be an IGMP setting if I recall correctly. Second, what does it say exactly it show on the sender and reciever screens concerning the multicast- G4L does have the option to receive udpcast, and checking the lines, with lzop compression it is `lzop -d -c -` and with gzip it shows `gzip -c - ` but that is with linux, and it might just be the difference in how windows does it. So, it could just be the missing -. Being the mainter of G4L, it make use of udpcast every week. I generally make updates to my lab machines, and create a new g4l backup image, and the use udpcast to image all the other machines in the lab. So, one ftp download to a sinfle machine, and then multi-cast to all the others. I created a custome udpcast send disk, and 7 copies of a udp-recieve disk. Just put the sender disk in the one machine, and it has all options set, and is ready to go. Then boot other machines from the reciever disk, and then hit enter on any machine went all are ready. I setup these CDs to do a whole disk image, so no inputed needed at all. Works just great. Another options. In my lab, the XP partition is always get messed up or changes, and we were recently getting this kavo virus being spread by flahes, and Norton was not stopping it at all. In this case, my XP partition imagge is about 6GB, so I just copied it to the Linux partition, which is 50% of the 80GB drive. So, it had plenty of free space. I then have it do a local ntfsclone restore to the XP partition, and can reimage the windows in about 10 minutes. NTFSCLONE is faster, since it only copies used data, and by not downloadin the image to all machines at once puts no load on the network. So, thre are some things to look at. +----------------------------------------------------------+ Michael D. Setzer II - Computer Science Instructor Guam Community College Computer Center mailto:mikes at kuentos.guam.net mailto:msetzerii at gmail.com http://www.guam.net/home/mikes Guam - Where America's Day Begins +----------------------------------------------------------+ http://setiathome.berkeley.edu (Original) Number of Seti Units Returned: 19,471 Processing time: 32 years, 290 days, 12 hours, 58 minutes (Total Hours: 287,489) BOINC at HOME CREDITS SETI 4,925,706.239266 | EINSTEIN 1,464,634.787568 | ROSETTA 428,541.225917 From kari.hyvonen at pp3.inet.fi Tue Mar 18 20:30:12 2008 From: kari.hyvonen at pp3.inet.fi (=?ISO-8859-1?Q?Kari_Hyv=F6nen?=) Date: Tue, 18 Mar 2008 21:30:12 +0200 Subject: [Udpcast] Rogue packets? Message-ID: <47E01844.1020500@pp3.inet.fi> We have used udpcast for a while at our school and it has performed flawlessly. Recently, we have cloned succesfully a bunch of Dell Optiplex 745 machines. Now, at another site, cloning is erratic. Most receiving machines get "rogue packet" messages, and on these machines the cloning seems to "hang". However, some machines finish ok. Is this a network switch problem at the site or something else? Kari Hyvönen Kerava, Finland From stefan.jaeckel at wiwi.uni-halle.de Tue Mar 25 08:25:10 2008 From: stefan.jaeckel at wiwi.uni-halle.de (Stefan =?iso-8859-1?q?J=E4ckel?=) Date: Tue, 25 Mar 2008 08:25:10 +0100 Subject: [Udpcast] Rogue packets? In-Reply-To: <47E01844.1020500@pp3.inet.fi> References: <47E01844.1020500@pp3.inet.fi> Message-ID: <200803250825.10577.stefan.jaeckel@wiwi.uni-halle.de> Am Dienstag 18 März 2008 20:30:12 schrieb Kari Hyvönen: > We have used udpcast for a while at our school and it has performed > flawlessly. Recently, we have cloned succesfully a bunch of Dell > Optiplex 745 machines. Now, at another site, cloning is erratic. Most > receiving machines get "rogue packet" messages, and on these machines > the cloning seems to "hang". However, some machines finish ok. Is this a > network switch problem at the site or something else? > > Kari Hyvönen > Kerava, Finland > _______________________________________________ > Udpcast mailing list > Udpcast at udpcast.linux.lu > https://lll.lgl.lu/mailman/listinfo/udpcast I had the same problem, with more than one udp-sender/receiver processes over "one wire". Maybe an other multicast application at the same time? I have a switched network. Stefan From karhyv at dlc.fi Tue Mar 25 18:34:25 2008 From: karhyv at dlc.fi (=?ISO-8859-1?Q?Kari_Hyv=F6nen?=) Date: Tue, 25 Mar 2008 19:34:25 +0200 Subject: [Udpcast] Udpcast Digest, Vol 57, Issue 5 In-Reply-To: References: Message-ID: <47E937A1.6050805@dlc.fi> udpcast-request at udpcast.linux.lu kirjoitti: >> We have used udpcast for a while at our school and it has performed >> flawlessly. Recently, we have cloned succesfully a bunch of Dell >> Optiplex 745 machines. Now, at another site, cloning is erratic. Most >> receiving machines get "rogue packet" messages, and on these machines >> the cloning seems to "hang". However, some machines finish ok. Is this a >> network switch problem at the site or something else? > I had the same problem, with more than one udp-sender/receiver processes > over "one wire". Maybe an other multicast application at the same time? I > have a switched network. > > Stefan At our site one the receiving workstations turned out to be faulty. It may have caused the error - the exact mechanism would be nice to know. ;) From tomc at bio.umass.edu Thu Mar 27 21:12:06 2008 From: tomc at bio.umass.edu (Tom Carpenter) Date: Thu, 27 Mar 2008 16:12:06 -0400 Subject: [Udpcast] compiling udpcast on Solaris (SPARC)? Message-ID: <47EBFF96.4040906@bio.umass.edu> Has anyone succeeded in compiling udpcast under Solaris 9 (SPARC)? I've included entries from config.log below. With respect to the stdint.h errors I found the following (http://www.squid-cache.org/mail-archive/squid-dev/200305/0146.html) "`inttypes.h' vs. `stdint.h' Paul Eggert notes that: ISO C 1999 says that `inttypes.h' includes `stdint.h', so there's no need to include `stdint.h' separately in a standard environment. Many implementations have `inttypes.h' but not `stdint.h' (e.g., Solaris 7), but I don't know of any implementation that has `stdint.h' but not `inttypes.h'. Nor do I know of any free software that includes `stdint.h'; `stdint.h' seems to be a creation of the committee." -Tom C. ================================================================== config.log entries ------------------ configure:3806: checking for stdint.h configure:3827: gcc -c -g -O2 conftest.c >&5 conftest.c:52:20: stdint.h: No such file or directory configure:3833: $? = 1 configure: failed program was: | /* confdefs.h. */ | #define PACKAGE_NAME "" | #define PACKAGE_TARNAME "" | #define PACKAGE_VERSION "" | #define PACKAGE_STRING "" | #define PACKAGE_BUGREPORT "" | #define HAVE_LIBPTHREAD 1 | #define STDC_HEADERS 1 | #define HAVE_SYS_WAIT_H 1 | #define HAVE_SYS_TYPES_H 1 | #define HAVE_SYS_STAT_H 1 | #define HAVE_STDLIB_H 1 | #define HAVE_STRING_H 1 | #define HAVE_MEMORY_H 1 | #define HAVE_STRINGS_H 1 | #define HAVE_INTTYPES_H 1 | /* end confdefs.h. */ | #include | #ifdef HAVE_SYS_TYPES_H | # include | #endif | #ifdef HAVE_SYS_STAT_H | # include | #endif | #ifdef STDC_HEADERS | # include | # include | #else | # ifdef HAVE_STDLIB_H | # include | # endif | #endif | #ifdef HAVE_STRING_H | # if !defined STDC_HEADERS && defined HAVE_MEMORY_H | # include | # endif | # include | #endif | #ifdef HAVE_STRINGS_H | # include | #endif | #ifdef HAVE_INTTYPES_H | # include | #endif | #ifdef HAVE_STDINT_H | # include | #endif | #ifdef HAVE_UNISTD_H | # include | #endif | | #include configure:3849: result: no [...SNIP...] configure:3894: checking net/if.h usability configure:3911: gcc -c -g -O2 conftest.c >&5 In file included from conftest.c:65: /usr/include/net/if.h:234: error: field `ifa_addr' has incomplete type /usr/include/net/if.h:236: error: field `ifu_broadaddr' has incomplete type /usr/include/net/if.h:237: error: field `ifu_dstaddr' has incomplete type /usr/include/net/if.h:269: error: field `lnr_addr' has incomplete type /usr/include/net/if.h:344: error: field `lifru_addr' has incomplete type /usr/include/net/if.h:345: error: field `lifru_dstaddr' has incomplete type /usr/include/net/if.h:346: error: field `lifru_broadaddr' has incomplete type /usr/include/net/if.h:347: error: field `lifru_token' has incomplete type /usr/include/net/if.h:348: error: field `lifru_subnet' has incomplete type /usr/include/net/if.h:387: error: field `sa_addr' has incomplete type /usr/include/net/if.h:397: error: field `slr_src' has incomplete type /usr/include/net/if.h:398: error: field `slr_grp' has incomplete type /usr/include/net/if.h:417: error: field `ifru_addr' has incomplete type /usr/include/net/if.h:418: error: field `ifru_dstaddr' has incomplete type /usr/include/net/if.h:420: error: field `ifru_broadaddr' has incomplete type /usr/include/net/if.h:488: error: parse error before "sa_family_t" /usr/include/net/if.h:491: error: parse error before '}' token /usr/include/net/if.h:501: error: parse error before "sa_family_t" /usr/include/net/if.h:510: error: parse error before '}' token /usr/include/net/if.h:641: error: field `ifta_saddr' has incomplete type /usr/include/net/if.h:642: error: field `ifta_daddr' has incomplete type configure:3917: $? = 1 configure: failed program was: | /* confdefs.h. */ | #define PACKAGE_NAME "" | #define PACKAGE_TARNAME "" | #define PACKAGE_VERSION "" | #define PACKAGE_STRING "" | #define PACKAGE_BUGREPORT "" | #define HAVE_LIBPTHREAD 1 | #define STDC_HEADERS 1 | #define HAVE_SYS_WAIT_H 1 | #define HAVE_SYS_TYPES_H 1 | #define HAVE_SYS_STAT_H 1 | #define HAVE_STDLIB_H 1 | #define HAVE_STRING_H 1 | #define HAVE_MEMORY_H 1 | #define HAVE_STRINGS_H 1 | #define HAVE_INTTYPES_H 1 | #define HAVE_UNISTD_H 1 | #define HAVE_FCNTL_H 1 | #define HAVE_LIMITS_H 1 | #define HAVE_SYS_IOCTL_H 1 | #define HAVE_SYS_TIME_H 1 | #define HAVE_ARPA_INET_H 1 | #define HAVE_NETDB_H 1 | #define HAVE_SYS_SELECT_H 1 | #define HAVE_SYS_SOCKIO_H 1 | #define HAVE_SYS_SOCKET_H 1 | #define HAVE_SYS_PARAM_H 1 | #define HAVE_MEMORY_H 1 | #define HAVE_MALLOC_H 1 | #define HAVE_SIGNAL_H 1 | /* end confdefs.h. */ | #include | #ifdef HAVE_SYS_TYPES_H | # include | #endif | #ifdef HAVE_SYS_STAT_H | # include | #endif | #ifdef STDC_HEADERS | # include | # include | #else | # ifdef HAVE_STDLIB_H | # include | # endif | #endif | #ifdef HAVE_STRING_H | # if !defined STDC_HEADERS && defined HAVE_MEMORY_H | # include | # endif | # include | #endif | #ifdef HAVE_STRINGS_H | # include | #endif | #ifdef HAVE_INTTYPES_H | # include | #endif | #ifdef HAVE_STDINT_H | # include | #endif | #ifdef HAVE_UNISTD_H | # include | #endif | #include configure:3931: result: no configure:3935: checking net/if.h presence configure:3950: gcc -E conftest.c configure:3956: $? = 0 configure:3970: result: yes configure:3983: WARNING: net/if.h: present but cannot be compiled configure:3985: WARNING: net/if.h: check for missing prerequisite headers? configure:3987: WARNING: net/if.h: see the Autoconf documentation configure:3989: WARNING: net/if.h: section "Present But Cannot Be Compiled" configure:3991: WARNING: net/if.h: proceeding with the preprocessor's result configure:3993: WARNING: net/if.h: in the future, the compiler will take precedence configure:3998: checking for net/if.h configure:4006: result: yes configure:3894: checking netinet/in.h usability configure:3911: gcc -c -g -O2 conftest.c >&5 configure:3917: $? = 0 configure:3931: result: yes