From john at jjdev.com Tue Mar 6 11:40:41 2007 From: john at jjdev.com (John de la Garza) Date: Tue, 6 Mar 2007 11:40:41 -0800 Subject: [buug] daylight savings changes Message-ID: <34AA9CFA-90CB-44CE-A864-0D2AE3D8DCC2@jjdev.com> I have a fedora core 3 box I need to update for the new daylight savings time From what I can tell I can just run: up2date tzdata to fix it then maybe system-config-date if it is needed is this correct? From jb at caustic.org Tue Mar 6 12:00:34 2007 From: jb at caustic.org (johan) Date: Tue, 06 Mar 2007 12:00:34 -0800 Subject: [buug] daylight savings changes In-Reply-To: <34AA9CFA-90CB-44CE-A864-0D2AE3D8DCC2@jjdev.com> References: <34AA9CFA-90CB-44CE-A864-0D2AE3D8DCC2@jjdev.com> Message-ID: <45EDC862.1020908@caustic.org> Check your tz information via zdump before you do anything that'll require system changes. Read the man pages on it first, but the basic command should be "zdump -c 2006,2008 -v /etc/localtime". Look to see if the change from PST to PDT happens on the 11th of March. John de la Garza wrote: > I have a fedora core 3 box I need to update for the new daylight savings > time > > > From what I can tell I can just run: > > up2date tzdata > > to fix it > > then maybe system-config-date if it is needed > > > is this correct? > _______________________________________________ > Buug mailing list > Buug at weak.org > http://www.weak.org/mailman/listinfo/buug From john at jjdev.com Tue Mar 6 12:04:17 2007 From: john at jjdev.com (John de la Garza) Date: Tue, 6 Mar 2007 12:04:17 -0800 Subject: [buug] daylight savings changes In-Reply-To: <45EDC862.1020908@caustic.org> References: <34AA9CFA-90CB-44CE-A864-0D2AE3D8DCC2@jjdev.com> <45EDC862.1020908@caustic.org> Message-ID: This system needs to be updated. I should have included this info, but I had already confirmed this. On Mar 6, 2007, at 12:00 PM, johan wrote: > Check your tz information via zdump before you do anything that'll > require system changes. > > Read the man pages on it first, but the basic command should be "zdump > -c 2006,2008 -v /etc/localtime". Look to see if the change from PST to > PDT happens on the 11th of March. > > John de la Garza wrote: >> I have a fedora core 3 box I need to update for the new daylight >> savings >> time >> >> >> From what I can tell I can just run: >> >> up2date tzdata >> >> to fix it >> >> then maybe system-config-date if it is needed >> >> >> is this correct? >> _______________________________________________ >> Buug mailing list >> Buug at weak.org >> http://www.weak.org/mailman/listinfo/buug > On Mar 6, 2007, at 12:00 PM, johan wrote: > Check your tz information via zdump before you do anything that'll > require system changes. > > Read the man pages on it first, but the basic command should be "zdump > -c 2006,2008 -v /etc/localtime". Look to see if the change from PST to > PDT happens on the 11th of March. > > John de la Garza wrote: >> I have a fedora core 3 box I need to update for the new daylight >> savings >> time >> >> >> From what I can tell I can just run: >> >> up2date tzdata >> >> to fix it >> >> then maybe system-config-date if it is needed >> >> >> is this correct? >> _______________________________________________ >> Buug mailing list >> Buug at weak.org >> http://www.weak.org/mailman/listinfo/buug > From john at jjdev.com Tue Mar 6 12:17:16 2007 From: john at jjdev.com (John de la Garza) Date: Tue, 6 Mar 2007 12:17:16 -0800 Subject: [buug] daylight savings changes In-Reply-To: <45EDC8F1.8070604@baywinds.org> References: <34AA9CFA-90CB-44CE-A864-0D2AE3D8DCC2@jjdev.com> <45EDC8F1.8070604@baywinds.org> Message-ID: <1CA9E2C7-2038-4AEB-9238-A46FE4B9A126@jjdev.com> I like this better. Thanks a million. On Mar 6, 2007, at 12:02 PM, Bruce Ferrell wrote: > I'm not sure if FC3 is still being maintained, so here's a how-to > to bring your zone files up to date > > http://www.linux-watch.com/news/NS6300294422.html > > > John de la Garza wrote: >> I have a fedora core 3 box I need to update for the new daylight >> savings time >> From what I can tell I can just run: >> up2date tzdata >> to fix it >> then maybe system-config-date if it is needed >> is this correct? >> _______________________________________________ >> Buug mailing list >> Buug at weak.org >> http://www.weak.org/mailman/listinfo/buug From john at jjdev.com Wed Mar 7 13:52:50 2007 From: john at jjdev.com (John de la Garza) Date: Wed, 7 Mar 2007 13:52:50 -0800 Subject: [buug] dd to clone Message-ID: I have a linux box that is up and running in a production environment. I want to reproduce the box for a backup server. If I want to use dd to copy the harddrive could I simply do something like: dd if=/dev/hda bs=1024k of=/dev/hdb or must I know more about the hard drives and do some calculations for how much data to copy? hdb is greater than or equal to the size of hda in this case From bferrell at baywinds.org Wed Mar 7 15:09:20 2007 From: bferrell at baywinds.org (Bruce Ferrell) Date: Wed, 07 Mar 2007 15:09:20 -0800 Subject: [buug] dd to clone In-Reply-To: References: Message-ID: <45EF4620.4020301@baywinds.org> use tar. it'll give you an accurate copy independant of the underlying hardware. something like this: tar cvf - | (cd ; tar xvf -) John de la Garza wrote: > I have a linux box that is up and running in a production environment. > I want to reproduce the box for a backup server. If I want to use dd > to copy the harddrive could I simply do something like: > > dd if=/dev/hda bs=1024k of=/dev/hdb > > or must I know more about the hard drives and do some calculations for > how much data to copy? > > hdb is greater than or equal to the size of hda in this case > _______________________________________________ > Buug mailing list > Buug at weak.org > http://www.weak.org/mailman/listinfo/buug > > From bferrell at baywinds.org Wed Mar 7 17:49:00 2007 From: bferrell at baywinds.org (Bruce Ferrell) Date: Wed, 07 Mar 2007 17:49:00 -0800 Subject: [buug] dd to clone In-Reply-To: <5d6a96c10703071633h65ef9dd8k40634bd64e8bbcbd@mail.gmail.com> References: <45EF4620.4020301@baywinds.org> <5d6a96c10703071633h65ef9dd8k40634bd64e8bbcbd@mail.gmail.com> Message-ID: <45EF6B8C.1010407@baywinds.org> I'd have to look up the switches to tell you how to make it correctly handle hard and sym links Anton Stroganov wrote: > Does tar handle symlinks and hardlinks through? > > On 3/7/07, Bruce Ferrell > wrote: > > use tar. it'll give you an accurate copy independant of the underlying > hardware. something like this: > > tar cvf - | (cd ; tar xvf -) > > > > John de la Garza wrote: > > I have a linux box that is up and running in a production > environment. > > I want to reproduce the box for a backup server. If I want to > use dd > > to copy the harddrive could I simply do something like: > > > > dd if=/dev/hda bs=1024k of=/dev/hdb > > > > or must I know more about the hard drives and do some > calculations for > > how much data to copy? > > > > hdb is greater than or equal to the size of hda in this case > > _______________________________________________ > > Buug mailing list > > Buug at weak.org > > http://www.weak.org/mailman/listinfo/buug > > > > > _______________________________________________ > Buug mailing list > Buug at weak.org > http://www.weak.org/mailman/listinfo/buug > > From Michael.Paoli at cal.berkeley.edu Wed Mar 7 20:36:27 2007 From: Michael.Paoli at cal.berkeley.edu (Michael Paoli) Date: Wed, 7 Mar 2007 20:36:27 -0800 Subject: [buug] dd to clone In-Reply-To: References: Message-ID: <1173328587.45ef92cb6fdea@webmail.rawbw.com> Quoting John de la Garza : > I have a linux box that is up and running in a production environment. > I want to reproduce the box for a backup server. If I want to use dd > to copy the harddrive could I simply do something like: > dd if=/dev/hda bs=1024k of=/dev/hdb > or must I know more about the hard drives and do some calculations for > how much data to copy? > hdb is greater than or equal to the size of hda in this case I've written a fair bit around this general topic area. One might want to peek at., e.g. these USENET articles for at least a bit of an overview of some of the considerations and methods: news:1137906739.932912.234000 at g43g2000cwa.googlegroups.com news:1147144043.547872.221020 at j73g2000cwa.googlegroups.com From john at jjdev.com Thu Mar 8 14:14:02 2007 From: john at jjdev.com (John de la Garza) Date: Thu, 8 Mar 2007 14:14:02 -0800 Subject: [buug] dd to clone In-Reply-To: <45EF4620.4020301@baywinds.org> References: <45EF4620.4020301@baywinds.org> Message-ID: <8C66635D-8F84-4838-B266-85008EC35433@jjdev.com> This wont copy the MBR or the partition table. I am looking for a way to clone a harddrive, not just a directory. On Mar 7, 2007, at 3:09 PM, Bruce Ferrell wrote: > use tar. it'll give you an accurate copy independant of the > underlying hardware. something like this: > > tar cvf - | (cd ; tar xvf -) > > > > John de la Garza wrote: >> I have a linux box that is up and running in a production >> environment. >> I want to reproduce the box for a backup server. If I want to use dd >> to copy the harddrive could I simply do something like: >> dd if=/dev/hda bs=1024k of=/dev/hdb >> or must I know more about the hard drives and do some calculations >> for >> how much data to copy? >> hdb is greater than or equal to the size of hda in this case >> _______________________________________________ >> Buug mailing list >> Buug at weak.org >> http://www.weak.org/mailman/listinfo/buug From bferrell at baywinds.org Thu Mar 8 14:17:04 2007 From: bferrell at baywinds.org (Bruce Ferrell) Date: Thu, 08 Mar 2007 14:17:04 -0800 Subject: [buug] dd to clone In-Reply-To: <8C66635D-8F84-4838-B266-85008EC35433@jjdev.com> References: <45EF4620.4020301@baywinds.org> <8C66635D-8F84-4838-B266-85008EC35433@jjdev.com> Message-ID: <45F08B60.2000006@baywinds.org> true, no mbr. got something different: http://air-imager.sourceforge.net/ Just saw it today John de la Garza wrote: > This wont copy the MBR or the partition table. I am looking for a way > to clone a harddrive, not just a directory. > > On Mar 7, 2007, at 3:09 PM, Bruce Ferrell wrote: > >> use tar. it'll give you an accurate copy independant of the >> underlying hardware. something like this: >> >> tar cvf - | (cd ; tar xvf -) >> >> >> >> John de la Garza wrote: >> >>> I have a linux box that is up and running in a production environment. >>> I want to reproduce the box for a backup server. If I want to use dd >>> to copy the harddrive could I simply do something like: >>> dd if=/dev/hda bs=1024k of=/dev/hdb >>> or must I know more about the hard drives and do some calculations for >>> how much data to copy? >>> hdb is greater than or equal to the size of hda in this case >>> _______________________________________________ >>> Buug mailing list >>> Buug at weak.org >>> http://www.weak.org/mailman/listinfo/buug > > > From john at jjdev.com Thu Mar 8 15:18:54 2007 From: john at jjdev.com (John de la Garza) Date: Thu, 8 Mar 2007 15:18:54 -0800 Subject: [buug] dd to clone In-Reply-To: <45F08B60.2000006@baywinds.org> References: <45EF4620.4020301@baywinds.org> <8C66635D-8F84-4838-B266-85008EC35433@jjdev.com> <45F08B60.2000006@baywinds.org> Message-ID: <23B8C63F-F149-4848-BC33-3B6C9177CFE7@jjdev.com> This kind of gets back to me wanting to use dd. From what I see this is a gui to dd (and some other things like netcat). I'm kind of in need of a command line tool. I only have a shell interface to work with. Although, if I play with this I may get some ideas from looking at what it does to image a drive. On Mar 8, 2007, at 2:17 PM, Bruce Ferrell wrote: > true, no mbr. got something different: > > http://air-imager.sourceforge.net/ > > Just saw it today > > John de la Garza wrote: >> This wont copy the MBR or the partition table. I am looking for >> a way to clone a harddrive, not just a directory. >> On Mar 7, 2007, at 3:09 PM, Bruce Ferrell wrote: >>> use tar. it'll give you an accurate copy independant of the >>> underlying hardware. something like this: >>> >>> tar cvf - | (cd ; tar xvf -) >>> >>> >>> >>> John de la Garza wrote: >>> >>>> I have a linux box that is up and running in a production >>>> environment. >>>> I want to reproduce the box for a backup server. If I want to >>>> use dd >>>> to copy the harddrive could I simply do something like: >>>> dd if=/dev/hda bs=1024k of=/dev/hdb >>>> or must I know more about the hard drives and do some >>>> calculations for >>>> how much data to copy? >>>> hdb is greater than or equal to the size of hda in this case >>>> _______________________________________________ >>>> Buug mailing list >>>> Buug at weak.org >>>> http://www.weak.org/mailman/listinfo/buug From bferrell at baywinds.org Thu Mar 8 15:29:18 2007 From: bferrell at baywinds.org (Bruce Ferrell) Date: Thu, 08 Mar 2007 15:29:18 -0800 Subject: [buug] dd to clone In-Reply-To: <23B8C63F-F149-4848-BC33-3B6C9177CFE7@jjdev.com> References: <45EF4620.4020301@baywinds.org> <8C66635D-8F84-4838-B266-85008EC35433@jjdev.com> <45F08B60.2000006@baywinds.org> <23B8C63F-F149-4848-BC33-3B6C9177CFE7@jjdev.com> Message-ID: <45F09C4E.60200@baywinds.org> I'm not too sure about putting the MBR from one disk onto a dissimilar disk. As I recall there may be stuff specific to the disk geometry... been a long time and I could be wrong there John de la Garza wrote: > This kind of gets back to me wanting to use dd. From what I see this > is a gui to dd (and some other things like netcat). I'm kind of in > need of a command line tool. I only have a shell interface to work with. > > Although, if I play with this I may get some ideas from looking at what > it does to image a drive. > > > On Mar 8, 2007, at 2:17 PM, Bruce Ferrell wrote: > >> true, no mbr. got something different: >> >> http://air-imager.sourceforge.net/ >> >> Just saw it today >> >> John de la Garza wrote: >> >>> This wont copy the MBR or the partition table. I am looking for a >>> way to clone a harddrive, not just a directory. >>> On Mar 7, 2007, at 3:09 PM, Bruce Ferrell wrote: >>> >>>> use tar. it'll give you an accurate copy independant of the >>>> underlying hardware. something like this: >>>> >>>> tar cvf - | (cd ; tar xvf -) >>>> >>>> >>>> >>>> John de la Garza wrote: >>>> >>>>> I have a linux box that is up and running in a production >>>>> environment. >>>>> I want to reproduce the box for a backup server. If I want to use dd >>>>> to copy the harddrive could I simply do something like: >>>>> dd if=/dev/hda bs=1024k of=/dev/hdb >>>>> or must I know more about the hard drives and do some >>>>> calculations for >>>>> how much data to copy? >>>>> hdb is greater than or equal to the size of hda in this case >>>>> _______________________________________________ >>>>> Buug mailing list >>>>> Buug at weak.org >>>>> http://www.weak.org/mailman/listinfo/buug > > > From togo at of.net Thu Mar 8 20:02:33 2007 From: togo at of.net (Tony Godshall) Date: Thu, 8 Mar 2007 20:02:33 -0800 Subject: [buug] dd to clone In-Reply-To: <45F09C4E.60200@baywinds.org> References: <45EF4620.4020301@baywinds.org> <8C66635D-8F84-4838-B266-85008EC35433@jjdev.com> <45F08B60.2000006@baywinds.org> <23B8C63F-F149-4848-BC33-3B6C9177CFE7@jjdev.com> <45F09C4E.60200@baywinds.org> Message-ID: <3424dc900703082002v5bb3a88ci2c6b8aaf34002b4e@mail.gmail.com> It usually works for me, as long as I don't go over those odd historical boundarys (like head/track/cyl-to-LBA or or 2GB or 8GB or old-LBA-to-48-bit). Forgive me if I'm fuzzy on the specifics- the trick seems to be that the partition tables are compatible. I often save off images for future use too, like this... # make sure empty space is compressible mount /dev/hdx1 /mnt dd if=/dev/zero of=/mnt/tmpzero; rm -f /mnt/tmpzero umount /mnt #(repeat for other partitions) cat /dev/hdx | gzip > /bigexternaldrive/hdx.img.gz On 3/8/07, Bruce Ferrell wrote: > I'm not too sure about putting the MBR from one disk onto a dissimilar > disk. As I recall there may be stuff specific to the disk geometry... > been a long time and I could be wrong there > > John de la Garza wrote: > > This kind of gets back to me wanting to use dd. From what I see this > > is a gui to dd (and some other things like netcat). I'm kind of in > > need of a command line tool. I only have a shell interface to work with. > > > > Although, if I play with this I may get some ideas from looking at what > > it does to image a drive. > > > > > > On Mar 8, 2007, at 2:17 PM, Bruce Ferrell wrote: > > > >> true, no mbr. got something different: > >> > >> http://air-imager.sourceforge.net/ > >> > >> Just saw it today > >> > >> John de la Garza wrote: > >> > >>> This wont copy the MBR or the partition table. I am looking for a > >>> way to clone a harddrive, not just a directory. > >>> On Mar 7, 2007, at 3:09 PM, Bruce Ferrell wrote: > >>> > >>>> use tar. it'll give you an accurate copy independant of the > >>>> underlying hardware. something like this: > >>>> > >>>> tar cvf - | (cd ; tar xvf -) > >>>> > >>>> > >>>> > >>>> John de la Garza wrote: > >>>> > >>>>> I have a linux box that is up and running in a production > >>>>> environment. > >>>>> I want to reproduce the box for a backup server. If I want to use dd > >>>>> to copy the harddrive could I simply do something like: > >>>>> dd if=/dev/hda bs=1024k of=/dev/hdb > >>>>> or must I know more about the hard drives and do some > >>>>> calculations for > >>>>> how much data to copy? > >>>>> hdb is greater than or equal to the size of hda in this case > >>>>> _______________________________________________ > >>>>> Buug mailing list > >>>>> Buug at weak.org > >>>>> http://www.weak.org/mailman/listinfo/buug > > > > > > > _______________________________________________ > Buug mailing list > Buug at weak.org > http://www.weak.org/mailman/listinfo/buug > -- -- Tony Godshall g From togo at of.net Thu Mar 8 20:20:14 2007 From: togo at of.net (Tony Godshall) Date: Thu, 8 Mar 2007 20:20:14 -0800 Subject: [buug] dd to clone In-Reply-To: <45F08B60.2000006@baywinds.org> References: <45EF4620.4020301@baywinds.org> <8C66635D-8F84-4838-B266-85008EC35433@jjdev.com> <45F08B60.2000006@baywinds.org> Message-ID: <3424dc900703082020n4ed181f8s87f6ca5a7c34fd53@mail.gmail.com> On 3/8/07, Bruce Ferrell wrote: > true, no mbr. got something different: > > http://air-imager.sourceforge.net/ > > Just saw it today ... Looks cool... a quick google search says these LiveCDs contain it. Helix, http://www.e-fense.com/helix/ BackTrack, http://www.remote-exploit.org/index.php/BackTrack 4BAK, http://4bak.org/ AliXe, http://alixe.org Any of them reputable? From togo at of.net Thu Mar 8 22:22:54 2007 From: togo at of.net (Tony Godshall) Date: Thu, 8 Mar 2007 22:22:54 -0800 Subject: [buug] dd to clone In-Reply-To: <3424dc900703082002v5bb3a88ci2c6b8aaf34002b4e@mail.gmail.com> References: <45EF4620.4020301@baywinds.org> <8C66635D-8F84-4838-B266-85008EC35433@jjdev.com> <45F08B60.2000006@baywinds.org> <23B8C63F-F149-4848-BC33-3B6C9177CFE7@jjdev.com> <45F09C4E.60200@baywinds.org> <3424dc900703082002v5bb3a88ci2c6b8aaf34002b4e@mail.gmail.com> Message-ID: <3424dc900703082222idb15631q281f693ed9cc2acd@mail.gmail.com> Oh, and I should mention that you can also dd just the MBR area, redo the partions on the new drive, mkfs then new partions, and copy the contents (I like rsync -avH). The Filesystems-HOWTO has a good section on imaging- looks like GNU parted is a single tool that can make the partitions right and copy (and even resize in place if you want to risk that). On 3/8/07, Tony Godshall wrote: > It usually works for me, as long as I don't go over those odd > historical boundarys (like head/track/cyl-to-LBA or or 2GB or 8GB or > old-LBA-to-48-bit). Forgive me if I'm fuzzy on the specifics- the > trick seems to be that the partition tables are compatible. > > I often save off images for future use too, like this... > > # make sure empty space is compressible > mount /dev/hdx1 /mnt > dd if=/dev/zero of=/mnt/tmpzero; rm -f /mnt/tmpzero > umount /mnt > #(repeat for other partitions) > cat /dev/hdx | gzip > /bigexternaldrive/hdx.img.gz > > On 3/8/07, Bruce Ferrell wrote: > > I'm not too sure about putting the MBR from one disk onto a dissimilar > > disk. As I recall there may be stuff specific to the disk geometry... > > been a long time and I could be wrong there > > > > John de la Garza wrote: > > > This kind of gets back to me wanting to use dd. From what I see this > > > is a gui to dd (and some other things like netcat). I'm kind of in > > > need of a command line tool. I only have a shell interface to work with. > > > > > > Although, if I play with this I may get some ideas from looking at what > > > it does to image a drive. > > > > > > > > > On Mar 8, 2007, at 2:17 PM, Bruce Ferrell wrote: > > > > > >> true, no mbr. got something different: > > >> > > >> http://air-imager.sourceforge.net/ > > >> > > >> Just saw it today > > >> > > >> John de la Garza wrote: > > >> > > >>> This wont copy the MBR or the partition table. I am looking for a > > >>> way to clone a harddrive, not just a directory. > > >>> On Mar 7, 2007, at 3:09 PM, Bruce Ferrell wrote: > > >>> > > >>>> use tar. it'll give you an accurate copy independant of the > > >>>> underlying hardware. something like this: > > >>>> > > >>>> tar cvf - | (cd ; tar xvf -) > > >>>> > > >>>> > > >>>> > > >>>> John de la Garza wrote: > > >>>> > > >>>>> I have a linux box that is up and running in a production > > >>>>> environment. > > >>>>> I want to reproduce the box for a backup server. If I want to use dd > > >>>>> to copy the harddrive could I simply do something like: > > >>>>> dd if=/dev/hda bs=1024k of=/dev/hdb > > >>>>> or must I know more about the hard drives and do some > > >>>>> calculations for > > >>>>> how much data to copy? > > >>>>> hdb is greater than or equal to the size of hda in this case > > >>>>> _______________________________________________ > > >>>>> Buug mailing list > > >>>>> Buug at weak.org > > >>>>> http://www.weak.org/mailman/listinfo/buug > > > > > > > > > > > _______________________________________________ > > Buug mailing list > > Buug at weak.org > > http://www.weak.org/mailman/listinfo/buug > > > > > -- > -- > Tony Godshall > g > -- -- Tony Godshall g From Michael.Paoli at cal.berkeley.edu Sat Mar 10 09:48:47 2007 From: Michael.Paoli at cal.berkeley.edu (Michael Paoli) Date: Sat, 10 Mar 2007 09:48:47 -0800 Subject: [buug] daylight saving changes, UNIX/LINUX/... time, etc. In-Reply-To: <1CA9E2C7-2038-4AEB-9238-A46FE4B9A126@jjdev.com> References: <34AA9CFA-90CB-44CE-A864-0D2AE3D8DCC2@jjdev.com> <45EDC8F1.8070604@baywinds.org> <1CA9E2C7-2038-4AEB-9238-A46FE4B9A126@jjdev.com> Message-ID: <1173548927.45f2ef7f94322@webmail.rawbw.com> Quoting John de la Garza: > I like this better. Thanks a million. > On Mar 6, 2007, at 12:02 PM, Bruce Ferrell wrote: > > I'm not sure if FC3 is still being maintained, so here's a how-to > > to bring your zone files up to date > > http://www.linux-watch.com/news/NS6300294422.html > > John de la Garza wrote: > >> I have a fedora core 3 box I need to update for the new daylight > >> savings time > >> From what I can tell I can just run: > >> up2date tzdata > >> to fix it > >> then maybe system-config-date if it is needed > >> is this correct? A fairly good article, > > http://www.linux-watch.com/news/NS6300294422.html with a lot of good information, ... but it gets a lot of details not exactly correct. Fist of all, the history of DST isn't that smooth and uniform. This change (enacted in 2005, taking effect starting in 2007) isn't the first time the DST rules/law(s) have significantly changed ... and not even the first significant changes since Uniform Time Act became effective in 1966. http://en.wikipedia.org/wiki/History_of_time_in_the_United_States#History_of_DST_in_the_US Also, the new changes for DST aren't necessarily permanent. Not only is there the issue of the legislature potentially changing things, but the Energy Policy act of 2005 contains explicit mention of "Right to Revert": " SEC. 110. DAYLIGHT SAVINGS. (a) Amendment.--Section 3(a) of the Uniform Time Act of 1966 (15 U.S.C. 260a(a)) is amended-- (1) by striking ``first Sunday of April'' and inserting ``second Sunday of March''; and (2) by striking ``last Sunday of October'' and inserting ``first Sunday of November''. (b) Effective Date.--Subsection (a) <> shall take effect 1 year after the date of enactment of this Act or March 1, 2007, whichever is later. (c) Report to Congress.--Not <> later than 9 months after the effective date stated in subsection (b), the Secretary shall report to Congress on the impact of this section on energy consumption in the United States. (d) Right to Revert.--Congress retains the right to revert the Daylight Saving Time back to the 2005 time schedules once the Department study is complete. " http://frwebgate.access.gpo.gov/cgi-bin/getdoc.cgi?dbname=109_cong_public_laws&docid=f:publ058.109 And on to LINUX, UNIX, times and clocks, etc. Historically, UNIX required a working real-time clock - specifically a hardware clock that would generally track the time even when the system was powered down (e.g. battery backed), or certainly at least would provide the correct time when the system was powered up - and before the operating system even began to boot. This clock would always be set to UTC. Over the years, particularly with different hardware, and LINUX, this "requirement" has often changed slightly. Some flavors of *nix (UNIX/LINUX/BSD/...) would boot - or at least start booting - even without a real-time hardware clock. They would generally require manual intervention (prompt for correct time) early in the boot-up sequence, and would not boot until that data was entered (typically prompted for on the console). Some *nix flavors will fall back to that behavior at boot if they detect a failure in the real-time hardware clock. Some *nix flavors would also, when booting, at least initially use the time of last sync of the root filesystem to set the system time (not to be confused with the real-time hardware clock - I'll get to system time in a bit). They would typically do this before attempting to read the real-time hardware clock and/or prompting for a correct date/time at boot. While reading the sync time from the root filesystem is only a very crude approximation of what the current date/time actually is, it does have the advantage of avoiding the nastiness that tends to happen if the system is ever run with a system time that is earlier than a system time it has already used before. Hardware clock is UTC? Historically, but not necessarily any more. It depends on both hardware, and convention (and/or operating system flavor). Most notably, with the advent of x86 PC hardware, certain other relatively predominant operating systems had a convention of having the hardware clock set to local time, rather than UTC. Consequently, various *nix flavors have different UTC vs. local conventions for the real-time hardware clock. Some require the real-time hardware clock be set to local time, others can be configured to have it set to either UTC or local time, and there may be some that require it to be UTC (none specifically jump to mind, but it is possible). Most LINUX distributions allow either convention, many of them default to local; UTC can be quite advantageous for greater compatibility with UNIX and for managing systems distributed across multiple timezones. Conversely, local can be advantageous for dual (or more) boot systems, where there may be other operating systems present that require it be set to local time. System time. In *nix, the operating system has a system time. It's tracked in seconds since the epoch (excluding leap seconds, if I recall correctly). For *nix, the epoch is 1970-01-01 00:00:00 UTC. When the system is booted, the system time is set (typically via reading the real-time hardware clock). Once the system time has been set, for the most part, the time of the real-time clock, and the system time, are relatively independent. For most all purposes, the operating system uses the system time, and does not refer back to or change the real-time clock. To actually change/set the real-time clock (which typically doesn't need to be done too often - especially if it's fairly accurate and set to UTC) is accomplished by various means, depending on the *nix flavor. E.g. for UNIX, the date(1) command sets both the system time and the real-time hardware clock - and it makes the appropriate conversions and such so each clock is set to its proper time (seconds since the epoch, and local or UTC time, respectively), and for LINUX, there are distinct commands (e.g. hwclock(8)) which are used to set the hardware clock. We'll get to these conversions in a bit (that's where all the DST and timezone excitement and such comes in). DST, timezones, conversions, and all that excitement. The operating system mostly just deals with the system time - as far as the operating system is concerned, that defines what the current time is. To deal with more conventional time information (date, hour, minutes, seconds, timezone, etc.), *nix does conversions between such time data and seconds since the epoch. For non-ancient *nix, this is mostly covered via timezone files, and also a timezone definition (e.g. 'which one is *my* timezone?'). The timezone definition is handled slightly differently among *nix flavors. Some always use the TZ environment variable to specify the timezone, and if it's not present, the zone will be presumed to be UTC (or GMT0). For other flavors (e.g. probably most or all LINUX distributions), the default timezone is specified via a file (e.g. containing or linked to a timezone data file), and if TZ isn't set in the environment, the default will be the zone specified in that file. Note also that *nix systems will generally define a default local timezone - e.g. by setting TZ (such as in /etc/profile or similar) or setting some default file, so that by default, most or all processes will pick up the default local timezone, unless they (or a parent process of theirs) takes explicit action to select a different timezone (by setting TZ to something else or unsetting it). Most commonly, the local/default timezone is set to the timezone corresponding to where the system is physically located - but that's not always the case (e.g. organizations spanning multiple zones may "standardize" their systems upon a single "local" zone or use UTC, and/or systems that are rather to quite mobile may use a particular zone rather than quite regularly changing the default). So, ... what's most critical to *nix, for proper DST handling and the like, are these timezone files. These files define offsets and transitions relative to the epoch, for their applicable zone, and also cover the name of the zone (by the file's relative pathname) and also the name(s)/abbreviation(s) used by the zone for both standard time, and also for Daylight Saving Time / Summertime. For well behaved programs (recommended practice) on *nix, the programs do all their time conversions via standard library calls, which in turn use the timezone files to know how to convert from any time in any zone, to seconds since the epoch, and vice versa. If the timezone files are correct and current, and the libraries and such work properly, and the programs all follow this practice, then updating the timezone files is all that's needed for dealing with changes to DST rules/law (e.g. if your timezone files covering the US predate the Energy Act of 2005, they aren't correct for at least 2007). Unfortunately not all software is ideal, so there are other "gottchas". Historically, UNIX didn't have timezone files - it was all coded within libraries, and these libraries were even statically linked into programs. In such cases, the programs themselves had to be updated. Also, some other programs/applications may not follow current recommended practice regarding handling of timezones, DST, time/date conversions, etc. E.g. if the program has its own hardcoded idea of what the DST rules are, and it's not coded to cover the more current stuff, the program will need to be updated. Some of the key areas to be aware of for such items on *nix, include Java, and at least some databases (many of which have their own time/date routines and coding). Note the variation in when Daylight/Standard time transitions have and will take place. If everything properly (directly or indirectly) does its calculations and conversions via proper timezone files, then this is all then well and fully covered by having a timezone file which is correct for past, present, and thus far know future rules/laws. $ zdump -v US/Pacific | sed -ne '/ 1970 /,/ Nov .* 2007 PST /{;s/^.* UTC = //p;}' Sun Apr 26 01:59:59 1970 PST isdst=0 gmtoff=-28800 Sun Apr 26 03:00:00 1970 PDT isdst=1 gmtoff=-25200 Sun Oct 25 01:59:59 1970 PDT isdst=1 gmtoff=-25200 Sun Oct 25 01:00:00 1970 PST isdst=0 gmtoff=-28800 Sun Apr 25 01:59:59 1971 PST isdst=0 gmtoff=-28800 Sun Apr 25 03:00:00 1971 PDT isdst=1 gmtoff=-25200 Sun Oct 31 01:59:59 1971 PDT isdst=1 gmtoff=-25200 Sun Oct 31 01:00:00 1971 PST isdst=0 gmtoff=-28800 Sun Apr 30 01:59:59 1972 PST isdst=0 gmtoff=-28800 Sun Apr 30 03:00:00 1972 PDT isdst=1 gmtoff=-25200 Sun Oct 29 01:59:59 1972 PDT isdst=1 gmtoff=-25200 Sun Oct 29 01:00:00 1972 PST isdst=0 gmtoff=-28800 Sun Apr 29 01:59:59 1973 PST isdst=0 gmtoff=-28800 Sun Apr 29 03:00:00 1973 PDT isdst=1 gmtoff=-25200 Sun Oct 28 01:59:59 1973 PDT isdst=1 gmtoff=-25200 Sun Oct 28 01:00:00 1973 PST isdst=0 gmtoff=-28800 Sun Jan 6 01:59:59 1974 PST isdst=0 gmtoff=-28800 Sun Jan 6 03:00:00 1974 PDT isdst=1 gmtoff=-25200 Sun Oct 27 01:59:59 1974 PDT isdst=1 gmtoff=-25200 Sun Oct 27 01:00:00 1974 PST isdst=0 gmtoff=-28800 Sun Feb 23 01:59:59 1975 PST isdst=0 gmtoff=-28800 Sun Feb 23 03:00:00 1975 PDT isdst=1 gmtoff=-25200 Sun Oct 26 01:59:59 1975 PDT isdst=1 gmtoff=-25200 Sun Oct 26 01:00:00 1975 PST isdst=0 gmtoff=-28800 Sun Apr 25 01:59:59 1976 PST isdst=0 gmtoff=-28800 Sun Apr 25 03:00:00 1976 PDT isdst=1 gmtoff=-25200 Sun Oct 31 01:59:59 1976 PDT isdst=1 gmtoff=-25200 Sun Oct 31 01:00:00 1976 PST isdst=0 gmtoff=-28800 Sun Apr 24 01:59:59 1977 PST isdst=0 gmtoff=-28800 Sun Apr 24 03:00:00 1977 PDT isdst=1 gmtoff=-25200 Sun Oct 30 01:59:59 1977 PDT isdst=1 gmtoff=-25200 Sun Oct 30 01:00:00 1977 PST isdst=0 gmtoff=-28800 Sun Apr 30 01:59:59 1978 PST isdst=0 gmtoff=-28800 Sun Apr 30 03:00:00 1978 PDT isdst=1 gmtoff=-25200 Sun Oct 29 01:59:59 1978 PDT isdst=1 gmtoff=-25200 Sun Oct 29 01:00:00 1978 PST isdst=0 gmtoff=-28800 Sun Apr 29 01:59:59 1979 PST isdst=0 gmtoff=-28800 Sun Apr 29 03:00:00 1979 PDT isdst=1 gmtoff=-25200 Sun Oct 28 01:59:59 1979 PDT isdst=1 gmtoff=-25200 Sun Oct 28 01:00:00 1979 PST isdst=0 gmtoff=-28800 Sun Apr 27 01:59:59 1980 PST isdst=0 gmtoff=-28800 Sun Apr 27 03:00:00 1980 PDT isdst=1 gmtoff=-25200 Sun Oct 26 01:59:59 1980 PDT isdst=1 gmtoff=-25200 Sun Oct 26 01:00:00 1980 PST isdst=0 gmtoff=-28800 Sun Apr 26 01:59:59 1981 PST isdst=0 gmtoff=-28800 Sun Apr 26 03:00:00 1981 PDT isdst=1 gmtoff=-25200 Sun Oct 25 01:59:59 1981 PDT isdst=1 gmtoff=-25200 Sun Oct 25 01:00:00 1981 PST isdst=0 gmtoff=-28800 Sun Apr 25 01:59:59 1982 PST isdst=0 gmtoff=-28800 Sun Apr 25 03:00:00 1982 PDT isdst=1 gmtoff=-25200 Sun Oct 31 01:59:59 1982 PDT isdst=1 gmtoff=-25200 Sun Oct 31 01:00:00 1982 PST isdst=0 gmtoff=-28800 Sun Apr 24 01:59:59 1983 PST isdst=0 gmtoff=-28800 Sun Apr 24 03:00:00 1983 PDT isdst=1 gmtoff=-25200 Sun Oct 30 01:59:59 1983 PDT isdst=1 gmtoff=-25200 Sun Oct 30 01:00:00 1983 PST isdst=0 gmtoff=-28800 Sun Apr 29 01:59:59 1984 PST isdst=0 gmtoff=-28800 Sun Apr 29 03:00:00 1984 PDT isdst=1 gmtoff=-25200 Sun Oct 28 01:59:59 1984 PDT isdst=1 gmtoff=-25200 Sun Oct 28 01:00:00 1984 PST isdst=0 gmtoff=-28800 Sun Apr 28 01:59:59 1985 PST isdst=0 gmtoff=-28800 Sun Apr 28 03:00:00 1985 PDT isdst=1 gmtoff=-25200 Sun Oct 27 01:59:59 1985 PDT isdst=1 gmtoff=-25200 Sun Oct 27 01:00:00 1985 PST isdst=0 gmtoff=-28800 Sun Apr 27 01:59:59 1986 PST isdst=0 gmtoff=-28800 Sun Apr 27 03:00:00 1986 PDT isdst=1 gmtoff=-25200 Sun Oct 26 01:59:59 1986 PDT isdst=1 gmtoff=-25200 Sun Oct 26 01:00:00 1986 PST isdst=0 gmtoff=-28800 Sun Apr 5 01:59:59 1987 PST isdst=0 gmtoff=-28800 Sun Apr 5 03:00:00 1987 PDT isdst=1 gmtoff=-25200 Sun Oct 25 01:59:59 1987 PDT isdst=1 gmtoff=-25200 Sun Oct 25 01:00:00 1987 PST isdst=0 gmtoff=-28800 Sun Apr 3 01:59:59 1988 PST isdst=0 gmtoff=-28800 Sun Apr 3 03:00:00 1988 PDT isdst=1 gmtoff=-25200 Sun Oct 30 01:59:59 1988 PDT isdst=1 gmtoff=-25200 Sun Oct 30 01:00:00 1988 PST isdst=0 gmtoff=-28800 Sun Apr 2 01:59:59 1989 PST isdst=0 gmtoff=-28800 Sun Apr 2 03:00:00 1989 PDT isdst=1 gmtoff=-25200 Sun Oct 29 01:59:59 1989 PDT isdst=1 gmtoff=-25200 Sun Oct 29 01:00:00 1989 PST isdst=0 gmtoff=-28800 Sun Apr 1 01:59:59 1990 PST isdst=0 gmtoff=-28800 Sun Apr 1 03:00:00 1990 PDT isdst=1 gmtoff=-25200 Sun Oct 28 01:59:59 1990 PDT isdst=1 gmtoff=-25200 Sun Oct 28 01:00:00 1990 PST isdst=0 gmtoff=-28800 Sun Apr 7 01:59:59 1991 PST isdst=0 gmtoff=-28800 Sun Apr 7 03:00:00 1991 PDT isdst=1 gmtoff=-25200 Sun Oct 27 01:59:59 1991 PDT isdst=1 gmtoff=-25200 Sun Oct 27 01:00:00 1991 PST isdst=0 gmtoff=-28800 Sun Apr 5 01:59:59 1992 PST isdst=0 gmtoff=-28800 Sun Apr 5 03:00:00 1992 PDT isdst=1 gmtoff=-25200 Sun Oct 25 01:59:59 1992 PDT isdst=1 gmtoff=-25200 Sun Oct 25 01:00:00 1992 PST isdst=0 gmtoff=-28800 Sun Apr 4 01:59:59 1993 PST isdst=0 gmtoff=-28800 Sun Apr 4 03:00:00 1993 PDT isdst=1 gmtoff=-25200 Sun Oct 31 01:59:59 1993 PDT isdst=1 gmtoff=-25200 Sun Oct 31 01:00:00 1993 PST isdst=0 gmtoff=-28800 Sun Apr 3 01:59:59 1994 PST isdst=0 gmtoff=-28800 Sun Apr 3 03:00:00 1994 PDT isdst=1 gmtoff=-25200 Sun Oct 30 01:59:59 1994 PDT isdst=1 gmtoff=-25200 Sun Oct 30 01:00:00 1994 PST isdst=0 gmtoff=-28800 Sun Apr 2 01:59:59 1995 PST isdst=0 gmtoff=-28800 Sun Apr 2 03:00:00 1995 PDT isdst=1 gmtoff=-25200 Sun Oct 29 01:59:59 1995 PDT isdst=1 gmtoff=-25200 Sun Oct 29 01:00:00 1995 PST isdst=0 gmtoff=-28800 Sun Apr 7 01:59:59 1996 PST isdst=0 gmtoff=-28800 Sun Apr 7 03:00:00 1996 PDT isdst=1 gmtoff=-25200 Sun Oct 27 01:59:59 1996 PDT isdst=1 gmtoff=-25200 Sun Oct 27 01:00:00 1996 PST isdst=0 gmtoff=-28800 Sun Apr 6 01:59:59 1997 PST isdst=0 gmtoff=-28800 Sun Apr 6 03:00:00 1997 PDT isdst=1 gmtoff=-25200 Sun Oct 26 01:59:59 1997 PDT isdst=1 gmtoff=-25200 Sun Oct 26 01:00:00 1997 PST isdst=0 gmtoff=-28800 Sun Apr 5 01:59:59 1998 PST isdst=0 gmtoff=-28800 Sun Apr 5 03:00:00 1998 PDT isdst=1 gmtoff=-25200 Sun Oct 25 01:59:59 1998 PDT isdst=1 gmtoff=-25200 Sun Oct 25 01:00:00 1998 PST isdst=0 gmtoff=-28800 Sun Apr 4 01:59:59 1999 PST isdst=0 gmtoff=-28800 Sun Apr 4 03:00:00 1999 PDT isdst=1 gmtoff=-25200 Sun Oct 31 01:59:59 1999 PDT isdst=1 gmtoff=-25200 Sun Oct 31 01:00:00 1999 PST isdst=0 gmtoff=-28800 Sun Apr 2 01:59:59 2000 PST isdst=0 gmtoff=-28800 Sun Apr 2 03:00:00 2000 PDT isdst=1 gmtoff=-25200 Sun Oct 29 01:59:59 2000 PDT isdst=1 gmtoff=-25200 Sun Oct 29 01:00:00 2000 PST isdst=0 gmtoff=-28800 Sun Apr 1 01:59:59 2001 PST isdst=0 gmtoff=-28800 Sun Apr 1 03:00:00 2001 PDT isdst=1 gmtoff=-25200 Sun Oct 28 01:59:59 2001 PDT isdst=1 gmtoff=-25200 Sun Oct 28 01:00:00 2001 PST isdst=0 gmtoff=-28800 Sun Apr 7 01:59:59 2002 PST isdst=0 gmtoff=-28800 Sun Apr 7 03:00:00 2002 PDT isdst=1 gmtoff=-25200 Sun Oct 27 01:59:59 2002 PDT isdst=1 gmtoff=-25200 Sun Oct 27 01:00:00 2002 PST isdst=0 gmtoff=-28800 Sun Apr 6 01:59:59 2003 PST isdst=0 gmtoff=-28800 Sun Apr 6 03:00:00 2003 PDT isdst=1 gmtoff=-25200 Sun Oct 26 01:59:59 2003 PDT isdst=1 gmtoff=-25200 Sun Oct 26 01:00:00 2003 PST isdst=0 gmtoff=-28800 Sun Apr 4 01:59:59 2004 PST isdst=0 gmtoff=-28800 Sun Apr 4 03:00:00 2004 PDT isdst=1 gmtoff=-25200 Sun Oct 31 01:59:59 2004 PDT isdst=1 gmtoff=-25200 Sun Oct 31 01:00:00 2004 PST isdst=0 gmtoff=-28800 Sun Apr 3 01:59:59 2005 PST isdst=0 gmtoff=-28800 Sun Apr 3 03:00:00 2005 PDT isdst=1 gmtoff=-25200 Sun Oct 30 01:59:59 2005 PDT isdst=1 gmtoff=-25200 Sun Oct 30 01:00:00 2005 PST isdst=0 gmtoff=-28800 Sun Apr 2 01:59:59 2006 PST isdst=0 gmtoff=-28800 Sun Apr 2 03:00:00 2006 PDT isdst=1 gmtoff=-25200 Sun Oct 29 01:59:59 2006 PDT isdst=1 gmtoff=-25200 Sun Oct 29 01:00:00 2006 PST isdst=0 gmtoff=-28800 Sun Mar 11 01:59:59 2007 PST isdst=0 gmtoff=-28800 Sun Mar 11 03:00:00 2007 PDT isdst=1 gmtoff=-25200 Sun Nov 4 01:59:59 2007 PDT isdst=1 gmtoff=-25200 Sun Nov 4 01:00:00 2007 PST isdst=0 gmtoff=-28800 $ From john at jjdev.com Thu Mar 15 11:15:00 2007 From: john at jjdev.com (John de la Garza) Date: Thu, 15 Mar 2007 11:15:00 -0700 Subject: [buug] clone a production server Message-ID: <616B31C0-7943-48BB-AD35-82C51E2CB44F@jjdev.com> A week or so ago I was asking how to use dd to back up a complete server. I think I should of stated what I needed and got feedback on how to best do it instead of wanting to use dd. I have a production server running a mod_perl web app that talks to Oracle. There are tons of symbolic and hard links and the documentation on how to build it is far from complete. I plan to reverse engineer it and use the current docs to one day (soon) understand it and be able to build it and clean up some of the clutter on it. The problem is, that if this box where to go down, we'd be hosed, so I want to copy it ASAP. What would be the best way to make a backup of the entire box as it is? Is this even really doable? I'd like to be able to make a hard drive that I could just switch with the one in the current server and have it work fine. The box isn't local, but at a data center. I can drive down there if I need to and I can shut it down, if need be. From grayarea at reddagger.org Thu Mar 15 12:35:52 2007 From: grayarea at reddagger.org (jwithers) Date: Thu, 15 Mar 2007 12:35:52 -0700 Subject: [buug] clone a production server In-Reply-To: <616B31C0-7943-48BB-AD35-82C51E2CB44F@jjdev.com> References: <616B31C0-7943-48BB-AD35-82C51E2CB44F@jjdev.com> Message-ID: <1173987352.5169.44.camel@localhost> Norton ghost if you don't want to do the whole dd dance. This is pretty much what it does. There is something called ghost for linux as well, but I haven't used it, so if it eats your hard drive, you didn't hear about it from me. You wanna get fancy, raid 1 the whole thing. http://www.tldp.org/HOWTO/Software-RAID-HOWTO-5.html#ss5.1 Which by definition will take care of your problem. Or, you could do it all manually with dd and parted, as has been suggested a couple of times I think. jpw On Thu, 2007-03-15 at 11:15 -0700, John de la Garza wrote: > A week or so ago I was asking how to use dd to back up a complete > server. I think I should of stated what I needed and got feedback on > how to best do it instead of wanting to use dd. > > I have a production server running a mod_perl web app that talks to > Oracle. There are tons of symbolic and hard links and the > documentation on how to build it is far from complete. I plan to > reverse engineer it and use the current docs to one day (soon) > understand it and be able to build it and clean up some of the > clutter on it. The problem is, that if this box where to go down, > we'd be hosed, so I want to copy it ASAP. > > What would be the best way to make a backup of the entire box as it > is? Is this even really doable? I'd like to be able to make a hard > drive that I could just switch with the one in the current server and > have it work fine. > > The box isn't local, but at a data center. I can drive down there > if I need to and I can shut it down, if need be. > > > _______________________________________________ > Buug mailing list > Buug at weak.org > http://www.weak.org/mailman/listinfo/buug -- jwithers From rick at linuxmafia.com Thu Mar 15 15:53:45 2007 From: rick at linuxmafia.com (Rick Moen) Date: Thu, 15 Mar 2007 15:53:45 -0700 Subject: [buug] clone a production server In-Reply-To: <616B31C0-7943-48BB-AD35-82C51E2CB44F@jjdev.com> References: <616B31C0-7943-48BB-AD35-82C51E2CB44F@jjdev.com> Message-ID: <20070315225345.GS28151@linuxmafia.com> Quoting John de la Garza (john at jjdev.com): > What would be the best way to make a backup of the entire box as it > is? [...] The box isn't local, but at a data center. Drive down with your favourite live-CD distribution (Knoppix will do) and a spare hard drive. Bring down the server. Connect the HD. Boot the live CD. Make filesystems on the spare HD to hold the filesystem replicas. Mount them. Mount (read-only) the filesystems from the production drive(s). Use your choice of file copying tool (I prefer rsync), as detailed here: "Copying Directory Trees" on http://linuxmafia.com/kb/Admin/ . Copying a running system's in-service filesystems is suboptimal. (Thus the suggestion of a live CD.) This averts a number of potential hazards, including filesytems changing while being copies, copying /proc accidentally (/proc/kcore's a bitch to copy ;-) ), copying other sub-mounted filesystems accidentally, etc. And dd's almost never the right tool, in my experience. -- Cheers, We write precisely We say exactly Rick Moen Since such is our habit in How to do a thing or how rick at linuxmafia.com Talking to machines; Every detail works. Excerpt from Prof. Touretzky's decss-haiku.txt @ http://www.cs.cmu.edu/~dst/ From lmth at deakin.edu.au Fri Mar 16 17:21:25 2007 From: lmth at deakin.edu.au (lmth at deakin.edu.au) Date: Sat, 17 Mar 2007 11:21:25 +1100 Subject: [buug] A request for your input. Message-ID: <20070317112125.mtnv1q81a8sosk0k@mail.deakin.edu.au> Hello My name is Lara Thynne and I am a PhD candidate at Deakin University Australia. I am currently researching the boundary between work and leisure activities directly related to the open source community and open source program development. As part of this I am running a survey at the following address. https://dcarf.deakin.edu.au/surveys/oss/ The survey is completely confidential and looks at your views and motivations to use Open Source software and to participate in the community. It will only take a five to ten minutes to complete and your contact details will not be recorded. You can withdraw your participation at any stage. I sincerely apologize for the spammish nature of this e-mail - I don't mean to abuse this list. I am trying to collect responses from as many open source developers and users as possible and a mailing list like can be the only way to reach many developers. Thanks again Lara P.S The program that I am using is open source, of course (www.phpsurveyor.org)! From paul.matthews at opensourcehowto.org Sat Mar 24 06:34:14 2007 From: paul.matthews at opensourcehowto.org (Paul Matthews) Date: Sat, 24 Mar 2007 23:34:14 +1000 (EST) Subject: [buug] OpenSourceHowTo.org Message-ID: <54647.124.187.80.146.1174743254.squirrel@www.opensourcehowto.org> hi everyone, i've setup a website so that users of open source server side software can come and look at instructional guides, how-to's, forums, wiki's and all sorts of other information about setting up software on Linux for both windows and Linux based networks. I am currently still working on it, adding new screen shots every day, hoping to double check a lot of my how-to's once the screen shots are added, re-write them to increase keyword density of my articles for better search engine results & once that is done i will hopefully be added streaming video of my how-to articles using either youtube embed code or revver embed code. http://www.opensourcehowto.org - OpenSourceHowTo.org http://www.opensourcehowto.org/how-to/openldap/setup-openldap.html - Setup OpenLDAP http://www.opensourcehowto.org/how-to/samba/openldap-lam-samba-as-pdc.html - OpenLDAP + LAM + Samba as PDC http://www.opensourcehowto.org/how-to/squid/squid-with-ntlm-authentication.html - NTLM authentication on squid http://www.opensourcehowto.org/how-to/squid/squid-with-pam-authentication.html - Squid and PAM authentication http://www.opensourcehowto.org/how-to/openldap/openssl--openldap.html - OpenLDAP and OpenSSL on 636 http://www.opensourcehowto.org/how-to/postfix/postfix-aliases-from-the-active-directory-cn.html - Postfix aliases from the Active Directory CN http://www.opensourcehowto.org/how-to/fedora/vsftpd--openssl--net2ftp.html - vsftpd + OpenSSL + Net2FTP http://www.opensourcehowto.org/how-to/squid/squid1-ntlm---dansguardian---squid2-cache.html - Squid1(ntlm) => Dansguardian => Squid2(cache) http://www.opensourcehowto.org/how-to/squid/squid-with-pam-authentication--squish-download-manager.html - Squid, pam authentication & Squish download manager http://www.opensourcehowto.org/how-to/squid/squid-and-havp.html - Squid and HAVP (http anti virus proxy) http://www.opensourcehowto.org/how-to/privoxy/privoxy--squid.html - Privoxy & Squid http://www.opensourcehowto.org/how-to/postfix/postfix--clamav--mailscanner--dovecot--ilohamail.html - Postfix + ClamAV + MailScanner + Dovercot http://www.opensourcehowto.org/how-to/fedora/installing-squidguard-on-fedora.html - Installing SquidGuard On Fedora http://www.opensourcehowto.org/how-to/dansguardian/dansguardian-with-different-filter-groups.html - Dansguardian with different filter groups http://www.opensourcehowto.org/how-to/apache/mambo--mysql--php--apache--ldap.html - Mambo + MySql + PHP + Apache http://www.opensourcehowto.org/how-to/fedora/opendc-hub--dcplusplus.html - OpenDC HUB & DCplusplus http://www.opensourcehowto.org/how-to/postfix/postfix--clamav--mailscanner--dovecot--ilohamail.html - Poptop VPN Server http://www.opensourcehowto.org/how-to/fedora/ddns-and-dhcp.html - DDNS using Bind9 and DHCP http://www.opensourcehowto.org/how-to/dovecot/pop3-server-on-fedora-with-ilohamail.html - pop3 Server On Fedora with IlohaMail http://www.opensourcehowto.org/how-to/ldap/linux-client-to-authenticate-against-ads-via-pam--ldap.html - LDAP linux client with OpenLDAP server http://www.opensourcehowto.org/how-to/ldap/linux-client-to-authenticate-against-ads-via-pam--ldap.html - LDAP linux client with Active Directory Server http://www.opensourcehowto.org/how-to/postfix/dovecot-imap--squirrel-mail--retrieve-user-data--active-directory--postfix.html - Dovecot, SquirrelMail, Retrieve User Data, Active Directory, Winbind, Postfix http://www.opensourcehowto.org/how-to/openldap/openldap-master-slave-replication.html - OpenLDAP Replication http://www.opensourcehowto.org/how-to/winbind/authentication-against-active-directories-using-winbind-for-pop3.html - Authentication against Active Directories using winbind for pop3 http://www.opensourcehowto.org/how-to/samba/swat-samba-web-administration-tool.html - SWAT (Samba Web AdministrationTool) http://www.opensourcehowto.org/how-to/postfix/openldap-and-postfix.html - OpenLDAP and postfix http://www.opensourcehowto.org/how-to/openldap/phpldapadmin--openldap.html - phpldapadmin and openldap http://www.opensourcehowto.org/how-to/fedora/xrdp--fedora-core-3.html - xrdp installation from source http://www.opensourcehowto.org/how-to/squid/personalized-denial-page-for-squid.html - Personalized Denial page for squid http://www.opensourcehowto.org/how-to/apache/setup-apache2-with-access-to-home-directories.html - Samba Primary Domain Controller with Group Policies http://www.opensourcehowto.org/how-to/apache/setup-apache2-with-access-to-home-directories.html - Setup Apache 2 with Access to Home Directories http://www.opensourcehowto.org/how-to/apache/setup-apache2-with-openldap-authentication.html - Setup Apache 2 with OpenLDAP Authentication http://www.opensourcehowto.org/how-to/fedora/setup-virtual-ip-address-on-eth01.html - Setup Virtual IP address on eth0:1 http://www.opensourcehowto.org/how-to/apache/setup-apache2-with-openssl.html - Setup Apache 2 with OpenSSL http://www.opensourcehowto.org/how-to/apache/apache2-with-webdav--htpasswd-using-openssl-certs.html - Apache 2 with Webdav & htpasswd using openSSL certs http://www.opensourcehowto.org/how-to/apache/apache2--virtual-hosts.html - Apache 2 & Virtual Hosts http://www.opensourcehowto.org/how-to/apache/apache-2.2-htpasswd--mod_authz_owner.html - apache 2.2 & mod_authz_owner http://www.opensourcehowto.org/how-to/apache/apache-with-user-access-to-their-own-home-directories-via-webdav-and-openldap-authentication.html - Apache, user access to home directories, webdav & openldap authentication http://www.opensourcehowto.org/how-to/fedora/tftp--pxe--syslinux--fedora-install.html - TFTP + PXE + syslinux + Fedora install http://www.opensourcehowto.org/how-to/dovecot/dovecot-imaps-pop3s-with-openssl-and-squirrelmail.html - Dovecot imap/pop3 with openssl and squirrelmail http://www.opensourcehowto.org/how-to/fedora/basic-nfs-share-mount-auto-mount.html - Basic NFS share, mount, auto mount http://www.opensourcehowto.org/how-to/postfix/postfix-with-procmail-vacation-auto-reply.html - Procmail Vacation Auto-Reply http://www.opensourcehowto.org/how-to/tftp/tftp--pxe--syslinux--g4l.html - TFTP + PXE + syslinux + G4L http://www.opensourcehowto.org/how-to/privoxy/setup-tor--privoxy-for-anonymous-searching.html - Setup Tor & Privoxy for anonymous searching http://www.opensourcehowto.org/how-to/mysql/squirrelmail-mysql-userprefs--mysql-address-book.html - SquirrelMail, Mysql userprefs & Mysql Address book http://www.opensourcehowto.org/how-to/postfix/postfix-with-openssl-and-saslauth.html - Postfix with OpenSSL and saslauth http://www.opensourcehowto.org/how-to/php/active-directory-as-address-book-with-apache-and-php.html - Active Directory as Address Book with apache and PHP http://www.opensourcehowto.org/how-to/winbind/winbind-virtual-users-postfix-spamassassin-procmail-squirrelmail--dovecot.html - Winbind Virtual Users, Postfix, Spamassassin, Procmail, SquirrelMail & Dovecot http://www.opensourcehowto.org/how-to/postfix/postfix-spamassassin-procmail-and-squirrelmail.html - Postfix, Spamassassin, Procmail and SquirrelMail http://www.opensourcehowto.org/how-to/fedora/print-server-for-a-windows-domain.html - Print Server for a windows domain with a web gui http://www.opensourcehowto.org/how-to/fedora/setup-virushammer.html - Virus scanning with VirusHammer http://www.opensourcehowto.org/how-to/apache/basic-cd-server-with-html-interface.html - Basic CD Server with HTML interface http://www.opensourcehowto.org/how-to/samba/deploy-firefox-with-samba-pdc--wpkg.html - deploy Firefox with Samba PDC & wpkg http://www.opensourcehowto.org/how-to/mysql/mysql-users-postfixadmin-postfix-dovecot--squirrelmail-with-userprefs-stored-in-mysql.html - MySQL users, PostfixAdmin, Postfix, Dovecot & SquirrelMail with userprefs stored in mySQL http://www.opensourcehowto.org/how-to/postfix/fighting-spam-with-spamassassin-pyzor-dcc-razor--rules-du-jour.html - Fighting Spam With SpamAssassin, Pyzor, DCC, Razor & Rules Du Jour http://www.opensourcehowto.org/how-to/mysql/backups-with-bacula-mysql--apache.html - backups with Bacula, MySql & Apache http://www.opensourcehowto.org/how-to/apache/backuppc--apache.html - BackupPc & Apache - VPN + ssl - Tomcat, MySql, J2ee, Alfresco CMS - Amanda backup - Squid & SARG - Squid with delay pools - Jabber messenger server with web client - Squid with proxy_auth & OpenLDAP - Samba file server with ClamAV doing on-access file scanning -- OpenSourceHowTo.org [url]http://www.opensourcehowto.org/[/url] Wiki.OpenSourceHowTo.org [url]http://wiki.opensourcehowto.org/[/url] My ServerSetup Scripts [url]http://evilperson85.110mb.com[/url] Please Support OpenSourceHowTo.org [url]http://www.opensourcehowto.org/how-to/welcome/support-opensourcehowto.org.html[/url] From itz at madbat.mine.nu Sun Mar 25 08:12:31 2007 From: itz at madbat.mine.nu (Ian Zimmerman) Date: 25 Mar 2007 11:12:31 -0400 Subject: [buug] apt ssh method In-Reply-To: <54647.124.187.80.146.1174743254.squirrel@www.opensourcehowto.org> Message-ID: <874po9we5c.fsf@unicorn.ahiker.homeip.net> Hello there :-) I am trying to use the ssh access method for apt, the Debian package tool. While this is mentioned in the sources.list(5) manpage, no examples of the URI syntax for this method are given, so I am left to guess. A quick google search was also unhelpful. I have tried the syntax below, and got the following result. deb ssh://my.remote.host debian/ Err ssh://my.remote.host debian/ Packages File not found ssh access as root via publickey to my.remote.host is configured and works. There's a ~root/debian/ directory on my.remote.host with the *.deb and Packages.gz files. -- Experience with Asset Control an asset. From grayarea at reddagger.org Sun Mar 25 12:45:07 2007 From: grayarea at reddagger.org (jwithers) Date: Sun, 25 Mar 2007 12:45:07 -0700 Subject: [buug] apt ssh method In-Reply-To: <874po9we5c.fsf@unicorn.ahiker.homeip.net> References: <874po9we5c.fsf@unicorn.ahiker.homeip.net> Message-ID: <1174851908.7081.617.camel@localhost> Ian, That slash at the end of the line source line really doesn't look right to me. I would think it should look something more like deb ssh://my.remote.host/ debian But I could be wrong, computers scare me and I try to learn as little about them as possible... jpw On Sun, 2007-03-25 at 11:12 -0400, Ian Zimmerman wrote: > Hello there :-) > > I am trying to use the ssh access method for apt, the Debian package tool. > While this is mentioned in the sources.list(5) manpage, no examples of the > URI syntax for this method are given, so I am left to guess. A quick google > search was also unhelpful. > > I have tried the syntax below, and got the following result. > > deb ssh://my.remote.host debian/ > > Err ssh://my.remote.host debian/ Packages > File not found > > > ssh access as root via publickey to my.remote.host is configured and works. > There's a ~root/debian/ directory on my.remote.host with the *.deb and Packages.gz > files. > -- jwithers From itz at madbat.mine.nu Sun Mar 25 16:41:26 2007 From: itz at madbat.mine.nu (Ian Zimmerman) Date: 25 Mar 2007 19:41:26 -0400 Subject: [buug] apt ssh method In-Reply-To: <1174851908.7081.617.camel@localhost> References: <874po9we5c.fsf@unicorn.ahiker.homeip.net> <1174851908.7081.617.camel@localhost> Message-ID: <87irco6gd5.fsf@unicorn.ahiker.homeip.net> Ian> I am trying to use the ssh access method for apt, the Debian Ian> package tool. While this is mentioned in the sources.list(5) Ian> manpage, no examples of the URI syntax for this method are given, Ian> so I am left to guess. A quick google search was also unhelpful. Ian> Ian> I have tried the syntax below, and got the following result. Ian> Ian> deb ssh://my.remote.host debian/ jpw> Ian, That slash at the end of the line source line really doesn't jpw> look right to me. I would think it should look something more like jpw> deb ssh://my.remote.host/ debian Actually, it turned out the other way - I needed _more_ slashes I had to put an absolute path there, i.e. deb ssh://my.remote.host /root/debian/ I decided to move the stuff out of root's homedir anyway, so it's just as well. In addition, I discovered what is arguably a bug: in the Package file the Filename fields must be absolute paths, while for "deb file:/..." archives they must be relative (ie. ./some_package.deb for the shallow case). So you can't serve the same archive both as deb file:/ and deb ssh:// . jpw> But I could be wrong, computers scare me and I try to learn as jpw> little about them as possible... Hmm, that seems a bit out of place in the BUUG ... has my sarcasm radar rusted in this cold climate? By the way, apologies for posting an answer to a totally unrelated topic initially. I lazily hit the Followup key to get the addressee filled, I remembered to delete the References header, but the In-Reply-To header was hidden by my UI so I left it in . I often get worked up about this in lists, so I should have heeded by own advice and avoid the Followup key when follow-up wasn't what I was doing. Oh yes, you have "localhost" in your Message-Id. Welcome to my spiffy autoresponder -- Experience with Asset Control an asset. From itz at madbat.mine.nu Tue Mar 27 20:02:50 2007 From: itz at madbat.mine.nu (Ian Zimmerman) Date: 27 Mar 2007 23:02:50 -0400 Subject: [buug] DHCP: override name server conditionally Message-ID: <87slbqysrp.fsf@unicorn.ahiker.homeip.net> With the ISC DHCP client version 3, I would like to change the name server, just as a "supersede domain-name-servers" line in dhclient.conf would do, but _only_ if I am talking to a particular DHCP server. The right place to change seems to be a script in dhclient-enter-hooks.d that redefines the shell function make_resolv_conf. The DHCP protocol option dhcp-server-identifier seems to be almost right for this, except that it contains the IP address which will be 192.168.0.1 in this case (and also in other cases), not helpful. Ideally I would have the MAC address of the server. Well, I could put an arp command in the script. But that is getting really too hackish. Hmm. Since I control this particular DHCP server, maybe I can make it return a recognizable hostname and use that as the condition. Any other suggestions? (Background: this is on a laptop that I use both at home and in other locations. My ISP's name servers are buggy, they demonstrably ignore TTLs in some cases and cache stuff much longer. So I want to avoid them.) -- Experience with Asset Control an asset. From Michael.Paoli at cal.berkeley.edu Wed Mar 28 09:02:43 2007 From: Michael.Paoli at cal.berkeley.edu (Michael Paoli) Date: Wed, 28 Mar 2007 08:02:43 -0800 Subject: [buug] DHCP: override name server conditionally In-Reply-To: <87slbqysrp.fsf@unicorn.ahiker.homeip.net> References: <87slbqysrp.fsf@unicorn.ahiker.homeip.net> Message-ID: <1175097763.460a91a382b86@webmail.rawbw.com> Quoting Ian Zimmerman : > With the ISC DHCP client version 3, I would like to change the name server, > just as a "supersede domain-name-servers" line in dhclient.conf would do, > but _only_ if I am talking to a particular DHCP server. > > The right place to change seems to be a script in dhclient-enter-hooks.d > that redefines the shell function make_resolv_conf. The DHCP protocol > option dhcp-server-identifier seems to be almost right for this, except > that it contains the IP address which will be 192.168.0.1 in this case > (and also in other cases), not helpful. Ideally I would have the MAC > address of the server. Well, I could put an arp command in the script. > But that is getting really too hackish. > > Hmm. Since I control this particular DHCP server, maybe I can make it > return a recognizable hostname and use that as the condition. > Any other suggestions? > > (Background: this is on a laptop that I use both at home and in other > locations. My ISP's name servers are buggy, they demonstrably ignore > TTLs in some cases and cache stuff much longer. So I want to avoid them.) Looking over at least some of the documentation, and keeping prudent programming practices in mind, etc., I come to at least approximately the same conclusions. At least for the ISC based DHCP client I have on Debian GNU/Linux 3.1, dhclient-script(8) seems to indicate the likely places to put such DHCP server specific client configurations. I agree that using an RFC 1918 IP address of server as the conditional (especially something as commonly used as 192.168.0.1) would be a poor choice of conditional. Determination by Ethernet MAC address might also be a less-than ideal selector - e.g. if the network changes (such as server gets replaced, or DHCP forwarding comes into play and server is no longer on the subnet), that may not be a good selector. What I'd suggest as a logical selector, would be the undesired DNS server IP addresses that are being handed out by certain DHCP server(s) - particularly if those are Internet IP addresses, and not RFC 1918 addresses. By using the non-desired IP addresses of DNS servers (which is the "problem" to begin with), triggering on that to put in the desired DNS server addresses, that would (or at least should?) reliably detect the undesired data, trigger on it, and in response configure with the desired data, rather than the undesired data. It would seem based upon dhclient-script(8) that something in/around make_resolv_conf would be the logical place to detect that conditional and adjust to take the corrective action. About the only probable downside I can see with that approach, is if the offending DHCP server switches to handing out different DNS server IP addresses that are also undesired, that event might not be quite as predictable. But that type of change probably doesn't occur frequently - and one might also broaden the check of offending DNS IP addresses (e.g. to all the IPs used/reserved for the ISP's infrastructure). Ethernet MAC addresses and/or IP address(es) of DHCP server may also potentially change - but again, those probably don't change all that often for the "offending" DHCP server. Another approach that might be possible, would be detecting some other parameters/data offered by the "offending" DHCP server that make it uniquely identifiable ... at least if such can be easily found and identified. Also, reverse DNS lookup of the offending (or potentially offending) DNS server IPs might be a pretty reliable check (e.g. if one looks them up against those DNS servers themselves, do they resolve to something in the domain of the offending ISP?). Anyway, hope that helps (at least a bit) ... might give one some useful ideas on various possible approaches and considerations, anyway. From itz at madbat.mine.nu Wed Mar 28 09:29:00 2007 From: itz at madbat.mine.nu (Ian Zimmerman) Date: 28 Mar 2007 12:29:00 -0400 Subject: [buug] IMP [Was: DHCP: override name server conditionally] In-Reply-To: <1175097763.460a91a382b86@webmail.rawbw.com> References: <87slbqysrp.fsf@unicorn.ahiker.homeip.net> <1175097763.460a91a382b86@webmail.rawbw.com> Message-ID: <87hcs58h83.fsf_-_@unicorn.ahiker.homeip.net> Interesting, I see that you use the IMP thing from horde. When I saw other people using it to post to lists, they always seemed to break threads (IIRC they had no References or In-Reply-To at all). It was such a pain that I have actually unsubscribed from some lists because of that. Yet I see yours works correctly. Care to share the secret so I can pass it on? Ian -- Experience with Asset Control an asset. From Michael.Paoli at cal.berkeley.edu Thu Mar 29 08:49:19 2007 From: Michael.Paoli at cal.berkeley.edu (Michael Paoli) Date: Thu, 29 Mar 2007 07:49:19 -0800 Subject: [buug] Re: IMP, horde & ISP [Was: Re: IMP [Was: DHCP: override ... In-Reply-To: <87hcs58h83.fsf_-_@unicorn.ahiker.homeip.net> References: <87slbqysrp.fsf@unicorn.ahiker.homeip.net> <1175097763.460a91a382b86@webmail.rawbw.com> <87hcs58h83.fsf_-_@unicorn.ahiker.homeip.net> Message-ID: <1175183359.460bdfff4a786@webmail.rawbw.com> Quoting Ian Zimmerman: > Interesting, I see that you use the IMP thing from horde. When I saw > other people using it to post to lists, they always seemed to break > threads (IIRC they had no References or In-Reply-To at all). It was > such a pain that I have actually unsubscribed from some lists because of > that. Yet I see yours works correctly. Care to share the secret so I > can pass it on? It's provided by my ISP. I've got an ISP: http://www.rawbandwidth.com/ that's substantially more competent and clueful (and also UNIX/LINUX/BSD/... friendly) than most. Costs a bit more, ... but also has *very* reasonable terms of service (a terse and perhaps slightly oversimplified version would probably be: "don't do illegal stuff, don't spam. For residential plans, no selling, giving away, or sharing of use of service outside of one's dwelling unit (that last restriction is lifted on the commercial plans)" ... that's about it. They also have a nice mix of offerings, including the set I want and have (rather than many of the 'large' consumer oriented ISPs, that mostly want to sell limited service to 'joe consumer' in a one-size-fits-all package (or a few such packages) which generally don't fit many of us. Oh, not that I am or would be pushing them for that, but they do also have a referral bonus (I've not gotten one of those yet, ... but one person did reap the benefits of one of those from my having signed up with Raw Bandwidth). They're not *perfect* ... but pretty darn close. As I mentioned, cost a bit more (but I think darn well worth it, and probably competitive for the service offerings, and level and quality of service), and it's not like there have been zero hiccups (but it's been so low, it's hardly worthy of mention - e.g. over about almost 4 years of service with them, I think they once had an outage of about 2 hours or so, ... and I only happened to know about it because they reported it to their customers (e-mailed us and explained what had happened, and duration and such) and reminded us about the dial-up that's available as a fallback at no charge whenever they have any DSL failure (their standard DSL packages also include a moderate number of hours (10?) of dial-up per month at no additional charge - and that can be upped to unlimited for a few $$ more per month ... and there's no counting of that time allocation when there's a DSL problem). I haven't checked, but last I recall hearing (and that might be out-of-date now) they're not offering and had no plans to offer support of IPv6 (I don't care about that *yet*, but probably will some day). And the only bug/problem I've found in the years with them so far, is I once found a very minor bug in their horde ... I e-mailed them about it, and they had it fixed in quite short order (I found it quite surprisingly fast of a correction at least compared to what I'd expect of most ISPs, ... I think they'd fixed it and e-mailed me back that it was fixed within a few hours or less, and perhaps it was even less or substantially less than an hour). Horde probably isn't my idea of ideal e-mail client, ... but with my ISP providing and maintaining it, and its mostly quite sufficient functionality it is quite convenient, and makes for one less thing for me to have to do much or anything about in terms of maintaining or configuring it. With the http/https access, it's also quite conveniently available from most any semi-sane or better browser that can generally talk to the Internet, even via proxies, where other protocols can't get through (e.g. blocked by firewalls/filters). So I tend to use it for "managing"/ handling my current and semi-current e-mail, ... and then I eventually suck down all the older stuff I want to keep (which is most of the stuff that isn't spam and isn't already archived elsewhere (such as many e-mail lists)) via fetchmail for my "archival" storage. From rick at linuxmafia.com Thu Mar 29 10:56:10 2007 From: rick at linuxmafia.com (Rick Moen) Date: Thu, 29 Mar 2007 10:56:10 -0700 Subject: [buug] Re: IMP, horde & ISP [Was: Re: IMP [Was: DHCP: override ... In-Reply-To: <1175183359.460bdfff4a786@webmail.rawbw.com> References: <87slbqysrp.fsf@unicorn.ahiker.homeip.net> <1175097763.460a91a382b86@webmail.rawbw.com> <87hcs58h83.fsf_-_@unicorn.ahiker.homeip.net> <1175183359.460bdfff4a786@webmail.rawbw.com> Message-ID: <20070329175609.GM32312@linuxmafia.com> Quoting Michael Paoli (Michael.Paoli at cal.berkeley.edu): > It's provided by my ISP. I've got an ISP: > http://www.rawbandwidth.com/ > that's substantially more competent and clueful (and also UNIX/LINUX/BSD/... > friendly) than most. Costs a bit more, ... but also has *very* reasonable > terms of service [...] Same here -- static-IP, home ADSL from Mike Durkin's Raw Bandwidth Communications. One of the many winning aspects of that company is that, any time SBC / PacBell / AT&T / whatever-name-du-jour's underlying frame transport shoots Mike's IP service in the foot, if only briefly (and this happens only very rarely), he's right on their case, tenaciously. If you're among the subscribers who called in to report such an outage, he keeps you briefed, and then sends out an explanatory report when things get fixed, and I swear to God he sounds _contrite_ about it. And upset at SBC for inconveniencing _his_ customers! I keep feeling I should send over some nice home-made baked goods with a note saying "Mike, relax! You're doing a phenomenal job." The amount of downtime is not even worth mentioning, over the many years I've been with Mike's firm -- and he even went out of his way to advise me on how to save money, when I moved service to our newer house, last year. I don't use any of the firm's services other than basic IP transport and occasionally one of the two large-provider Usenet services (Supernews and one other), but that's more than enough to earn high marks from me. From itz at madbat.mine.nu Thu Mar 29 10:58:06 2007 From: itz at madbat.mine.nu (Ian Zimmerman) Date: 29 Mar 2007 13:58:06 -0400 Subject: [buug] Archiving encrypted mail, how Message-ID: <87mz1w2aq9.fsf@unicorn.ahiker.homeip.net> About 3 years ago I started storing every email I ever send, and I think many people do that now. Well I started doing it with gmail, but since then a few things have happened such as Google cooperating with a dictatorship, so now I keep it on my own machine. I have my user agent auto-insert a Bcc header with my own address. A side benefit of doing it this way (rather than, e.g., just appending them to a file) is that threads in which I participate stay whole, even non-list threads. My problem/question is how to deal with PGP encrypted mail? Since the encryption, in this setup, happens before the mail is sent to any of its destinations, I get back the encrypted copy, which is useless. Does enigmail (sp?) handle this correctly? What about other MUAs? -- Experience with Asset Control an asset. From john at jjdev.com Fri Mar 30 13:21:07 2007 From: john at jjdev.com (John de la Garza) Date: Fri, 30 Mar 2007 13:21:07 -0700 Subject: [buug] arp Message-ID: <082D130C-058C-4AF6-9674-952F106F916A@jjdev.com> I have a webserver that is say ip addr: 1. I make change it to be 2, then make a new box (different nic) to be ip addr of 1. Now things behave odd... I just learned that things like this get cached. Is there a way I can erase the entire arp cache and let it get rebuilt? or must I do it case by case with arp -d? From itz at madbat.mine.nu Fri Mar 30 13:42:35 2007 From: itz at madbat.mine.nu (Ian Zimmerman) Date: 30 Mar 2007 16:42:35 -0400 Subject: [buug] arp In-Reply-To: <082D130C-058C-4AF6-9674-952F106F916A@jjdev.com> References: <082D130C-058C-4AF6-9674-952F106F916A@jjdev.com> Message-ID: <87mz1ua2f8.fsf@unicorn.ahiker.homeip.net> John> I have a webserver that is say ip addr: 1. I make change it to be John> 2, then make a new box (different nic) to be ip addr of 1. Now John> things behave odd... I just learned that things like this get John> cached. John> Is there a way I can erase the entire arp cache and let it get John> rebuilt? or must I do it case by case with arp -d? Maybe not what you asked .. but arp -f will have the same effect, provided has the right contents. -- Experience with Asset Control an asset. From jeremy at nirvani.net Fri Mar 30 15:29:37 2007 From: jeremy at nirvani.net (Jeremy Brand) Date: Fri, 30 Mar 2007 15:29:37 -0700 Subject: [buug] colocation space Message-ID: Howdy all, I just recently had to move colos for my self-owned business, and I took up residence in Hurricane Electric (fremont II) after leaving Market Post Tower in San Jose. I have one extra cabinet at Hurricane Electric (fmt2) and if anyone wants to drop in some *nix servers, I am hoping someone will help me offset my costs. BTW, no Windows servers welcome, thank you for asking. Here's my offer. - 12U of physical space in a shared 7foot locking, shared cabinet. - 5Amp of dedicated to you, breakered, UPS and generator backed power. - 1TB of data transfer per month. - Front and Rear mounting posts. (bring your own shelves [BYOS] or use rack-mount servers. - Unlimited IPs (your own subnet) with proper ARIN justification. And all the other standard features of a high-end colocation facility. I'm asking $200/month for this. Email me off list if interested. More details about Hurricane Electric at their website: www.he.net Sorry for the spam. Jeremy -- "Prediction is very difficult, especially of the future." Niels Bohr http://www.nirvani.net From Michael.Paoli at cal.berkeley.edu Sat Mar 31 20:49:29 2007 From: Michael.Paoli at cal.berkeley.edu (Michael Paoli) Date: Sat, 31 Mar 2007 19:49:29 -0800 Subject: [buug] arp In-Reply-To: <082D130C-058C-4AF6-9674-952F106F916A@jjdev.com> References: <082D130C-058C-4AF6-9674-952F106F916A@jjdev.com> Message-ID: <1175399369.460f2bc97a1da@webmail.rawbw.com> Quoting John de la Garza: > I have a webserver that is say ip addr: 1. I make change it to be 2, > then make a new box (different nic) to be ip addr of 1. Now things > behave odd... I just learned that things like this get cached. > > Is there a way I can erase the entire arp cache and let it get > rebuilt? or must I do it case by case with arp -d? One can generally check/alter the arp cache data via the arp command. Typically arp data will effectively fix itself (older data timing out) within several minutes. If the problem(s)/issue(s) are lasting longer than that, you likely have something else going on. /etc/ethers can be used for persistent Ethernet MAC <--> IP mapping (and thwarting arp cache poisoning) ... but that's typically not used in most environments. If you've still got persistent issues, you may want to poke around with tcpdump or the like, a bit more, to see what's actually going on on the network. With suitable options to tcpdump, one will get to see the arp request and reply packets (or note the lack thereof) and the Ethernet MAC addresses (link level headers). That should be enough to relatively quickly isolate if one has a Ethernet MAC address and/or arp or related issue/problem.