You can subscribe to this list here.
| 2002 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(15) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2003 |
Jan
(7) |
Feb
(6) |
Mar
(3) |
Apr
(9) |
May
(5) |
Jun
(11) |
Jul
(40) |
Aug
(15) |
Sep
(3) |
Oct
(2) |
Nov
(1) |
Dec
|
| 2004 |
Jan
(2) |
Feb
(8) |
Mar
(8) |
Apr
(25) |
May
(10) |
Jun
(11) |
Jul
(5) |
Aug
(9) |
Sep
(2) |
Oct
(7) |
Nov
(7) |
Dec
(6) |
| 2005 |
Jan
(6) |
Feb
(17) |
Mar
(9) |
Apr
(3) |
May
(4) |
Jun
(11) |
Jul
(42) |
Aug
(33) |
Sep
(13) |
Oct
(14) |
Nov
(19) |
Dec
(7) |
| 2006 |
Jan
(23) |
Feb
(19) |
Mar
(6) |
Apr
(8) |
May
(1) |
Jun
(12) |
Jul
(50) |
Aug
(16) |
Sep
(4) |
Oct
(18) |
Nov
(15) |
Dec
(10) |
| 2007 |
Jan
(10) |
Feb
(13) |
Mar
(1) |
Apr
|
May
(1) |
Jun
(1) |
Jul
(5) |
Aug
(7) |
Sep
(25) |
Oct
(58) |
Nov
(15) |
Dec
(4) |
| 2008 |
Jan
(2) |
Feb
(13) |
Mar
(3) |
Apr
(10) |
May
(1) |
Jun
(4) |
Jul
(4) |
Aug
(23) |
Sep
(21) |
Oct
(7) |
Nov
(3) |
Dec
(12) |
| 2009 |
Jan
(5) |
Feb
(2) |
Mar
|
Apr
(7) |
May
(6) |
Jun
(25) |
Jul
|
Aug
(2) |
Sep
(2) |
Oct
(3) |
Nov
(2) |
Dec
(2) |
| 2010 |
Jan
(1) |
Feb
|
Mar
(3) |
Apr
(2) |
May
(2) |
Jun
|
Jul
(3) |
Aug
(3) |
Sep
(2) |
Oct
|
Nov
|
Dec
(3) |
| 2011 |
Jan
(10) |
Feb
(2) |
Mar
(71) |
Apr
(4) |
May
(8) |
Jun
|
Jul
|
Aug
(4) |
Sep
(3) |
Oct
(1) |
Nov
(5) |
Dec
(1) |
| 2012 |
Jan
|
Feb
|
Mar
(2) |
Apr
(2) |
May
(15) |
Jun
(1) |
Jul
|
Aug
(20) |
Sep
|
Oct
|
Nov
(2) |
Dec
|
| 2013 |
Jan
(3) |
Feb
(5) |
Mar
|
Apr
(4) |
May
(2) |
Jun
(11) |
Jul
(12) |
Aug
|
Sep
(19) |
Oct
(25) |
Nov
(7) |
Dec
(9) |
| 2014 |
Jan
(6) |
Feb
(2) |
Mar
(7) |
Apr
(5) |
May
(1) |
Jun
|
Jul
(1) |
Aug
(1) |
Sep
(1) |
Oct
(3) |
Nov
(2) |
Dec
(4) |
| 2015 |
Jan
(8) |
Feb
(2) |
Mar
(3) |
Apr
(4) |
May
(3) |
Jun
|
Jul
(1) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(1) |
| 2016 |
Jan
(16) |
Feb
|
Mar
(1) |
Apr
(1) |
May
(12) |
Jun
(3) |
Jul
|
Aug
(5) |
Sep
|
Oct
(2) |
Nov
|
Dec
|
| 2017 |
Jan
(2) |
Feb
(6) |
Mar
(68) |
Apr
(18) |
May
(8) |
Jun
(1) |
Jul
|
Aug
(10) |
Sep
(2) |
Oct
(1) |
Nov
(13) |
Dec
(25) |
| 2018 |
Jan
(18) |
Feb
(2) |
Mar
|
Apr
(1) |
May
(1) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(3) |
Dec
|
| 2019 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
(1) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2020 |
Jan
|
Feb
(2) |
Mar
|
Apr
|
May
|
Jun
(1) |
Jul
|
Aug
(1) |
Sep
(1) |
Oct
(7) |
Nov
|
Dec
|
|
From: Craig B. <cba...@us...> - 2017-03-25 01:10:17
|
BackupPC community, I'm happy to announce that BackupPC 4.1.0 has been released on Github <https://github.com/backuppc/backuppc/releases>. There are also new versions of BackupPC::XS <https://github.com/backuppc/backuppc-xs/releases> (0.53) and rsync-bpc <https://github.com/backuppc/rsync-bpc/releases> (3.0.9.6). on Github. There are a few new features and important bug fixes, including: - BackupPC_migrateV3toV4 optionally allows old 3.x backups to be migrated to 4.x format, eliminating the hardlinks used by 3.x. (BackupPC 4.x can view/browse/restore 3.x backups, so it's not necessary to run BackupPC_migrateV3toV4) - modern CSS (from Ernesto Carrea) - fixes to rsync restore (from Stephen Joyce) - fixes to exponential backup expiry (from Alexander Moisseev) - added template systemd files - configure.pl adds a --config-only option for package builders (from Alexander Moisseev) If you are using 4.0.0, upgrading is strongly recommended. After installing those two packages, BackupPC 4.1.0 can be installed from the tar ball with: tar zxf BackupPC-4.1.0.tar.gz cd BackupPC-4.1.0 perl configure.pl See the README.md, ChangeLog and doc/BackupPC.html <http://backuppc.sourceforge.net/BackupPC-4.0.0.html> files for more information. An overview of the changes is below. Thanks for to everyone who provided fixes and reported issues. Enjoy! Craig * Merged pull requests: #17, #44, #59, #60, #61, #62,#68, #69, #73, #74 * Fixed certain rsync restores, based on patch from Stephen Joyce. * Improvements to Gentoo init.d script (#69) from sigmoidal. * Fixed exponential backup expiry algorithm, submitted by Alexander Moisseev (#17). * Added --config-only option to configure.pl, from Alexander Moisseev (#62). * Ensure $? is 0 in bin/BackupPC_dump UserCommandRun() if no command is run, reported and proposed fix by stirab (issue #67). * configure.pl replaces SourceForge links in $Conf{CgiNavBarLinks} with new Github links. * Added BackupPC_migrateV3toV4 that migrates old V3 backup storage to V4, eliminating hardlinks. The forward merging of V3 incrementals is not changed; just the backup tree is updated to use V4 attrib files, and the linked V3 files are removed. Thanks to Michael Huntley for testing. * Renamed init.d to systemd, added systemd/src/backuppc.service template * Replaced CSS file with new BackupPC_stnd.css from Ernesto Carrea (#59,#73). Renamed old ones to conf/BackupPC_retro_v2.css and conf/BackupPC_retro_v3.css. * Updated lib/BackupPC/DirOps.pm with more robust checking that IO::Dirent works. Matches similar changes to 3.x that didn't make it into 4.x. Fixes issue #56. * BackupPC_refCountUpdate handles case of empty backups better. It creates a noPoolCntOk file if there are legitimately no poolCnt files after running. Reported by Alexander Kobel. * Fixed umask() in child processes (issue #58) reported by Alexander Moisseev. |
|
From: Richard S. <hob...@gm...> - 2017-03-22 21:59:40
|
On Wed, Mar 22, 2017 at 4:56 PM, Kenneth Porter <sh...@se...> wrote: > On 3/22/2017 5:37 AM, Richard Shaw wrote: > > No, pretty sure that's not the case. There is a straightforward error > > in apache's error_log saying it's not allowing access due to the > > configuration and when I use a Directory directive to allow access to > > /usr/sbin it works, but I don't want to allow apache access to > > everything in /usr/sbin (or /usr/libexec) > > I thought you could put the binary in /usr/libexec/BackupPC and then use > a Directory directive on that. Hadn't thought of that... Thanks! Richard |
|
From: Kenneth P. <sh...@se...> - 2017-03-22 21:57:37
|
On 3/22/2017 5:37 AM, Richard Shaw wrote: > No, pretty sure that's not the case. There is a straightforward error > in apache's error_log saying it's not allowing access due to the > configuration and when I use a Directory directive to allow access to > /usr/sbin it works, but I don't want to allow apache access to > everything in /usr/sbin (or /usr/libexec) I thought you could put the binary in /usr/libexec/BackupPC and then use a Directory directive on that. --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus |
|
From: Steve P. <n9...@n9...> - 2017-03-22 15:26:45
|
Sorry, I re-posed this to the users list, it probably does not belong here. My apologies to all! :( Steve > On Mar 22, 2017, at 9:16 AM, Steve Palm <n9...@n9...> wrote: > > On Mar 4, 2017, at 1:12 PM, Craig Barratt <cba...@us... <mailto:cba...@us...>> wrote: >> I'm happy to announce that BackupPC 4.0.0 has been released on Github <https://github.com/backuppc/backuppc/releases> and SourceForge <https://sourceforge.net/projects/backuppc/files/backuppc/4.0.0/>. > > I was excited to read this, congratulations on a huge milestone! > >> BackupPC 4.0.0 requires the perl module BackupPC::XS <https://github.com/backuppc/backuppc-xs/releases> (>= 0.50) and rsync-bpc <https://github.com/backuppc/rsync-bpc/releases> (>= 3.0.9.5). > > I installed these. > >> After installing those two packages, BackupPC 4.0.0 can be installed from the tar ball with: >> >> tar zxf BackupPC-4.0.0.tar.gz >> cd BackupPC-4.0.0 >> perl configure.pl <http://configure.pl/> > And I was greeted with a very unhelpful.... > > BackupPC needs the package version. Please install version > before installing BackupPC. > > Even running with a trace is not helping me much, not sure what I'm looking at here: > > perl -d:Trace configure.pl > >> configure.pl:54: my @ConfigureBinList = qw( > >> configure.pl:79: my @ConfigureLibList = qw( > >> configure.pl:137: if ( $ConfigureBinList[0] eq "__" . "CONFIGURE_BIN_LIST__" ) { > >> configure.pl:145: my @Packages = qw(version Encode File::Path File::Spec File::Copy DirHandle > >> configure.pl:149: my $PackageVersion = { > >> configure.pl:154: foreach my $pkg ( @Packages ) { > >> configure.pl:155: eval "use $pkg"; > >> (eval 1)[configure.pl:155]:2: ;>> configure.pl:156: if ( !$@ ) { > >> configure.pl:169: if ( $pkg =~ /BackupPC::Lib/ ) { > >> configure.pl:184: die <<EOF; > > BackupPC needs the package version. Please install version > before installing BackupPC. > > Any ideas? > > Steve > > ------------------------------------------------------------------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot_______________________________________________ > BackupPC-devel mailing list > Bac...@li... > List: https://lists.sourceforge.net/lists/listinfo/backuppc-devel > Wiki: http://backuppc.wiki.sourceforge.net > Project: http://backuppc.sourceforge.net/ |
|
From: Steve P. <n9...@n9...> - 2017-03-22 14:48:59
|
On Mar 4, 2017, at 1:12 PM, Craig Barratt <cba...@us... <mailto:cba...@us...>> wrote: > I'm happy to announce that BackupPC 4.0.0 has been released on Github <https://github.com/backuppc/backuppc/releases> and SourceForge <https://sourceforge.net/projects/backuppc/files/backuppc/4.0.0/>. I was excited to read this, congratulations on a huge milestone! > BackupPC 4.0.0 requires the perl module BackupPC::XS <https://github.com/backuppc/backuppc-xs/releases> (>= 0.50) and rsync-bpc <https://github.com/backuppc/rsync-bpc/releases> (>= 3.0.9.5). I installed these. > After installing those two packages, BackupPC 4.0.0 can be installed from the tar ball with: > > tar zxf BackupPC-4.0.0.tar.gz > cd BackupPC-4.0.0 > perl configure.pl <http://configure.pl/> And I was greeted with a very unhelpful.... BackupPC needs the package version. Please install version before installing BackupPC. Even running with a trace is not helping me much, not sure what I'm looking at here: perl -d:Trace configure.pl >> configure.pl:54: my @ConfigureBinList = qw( >> configure.pl:79: my @ConfigureLibList = qw( >> configure.pl:137: if ( $ConfigureBinList[0] eq "__" . "CONFIGURE_BIN_LIST__" ) { >> configure.pl:145: my @Packages = qw(version Encode File::Path File::Spec File::Copy DirHandle >> configure.pl:149: my $PackageVersion = { >> configure.pl:154: foreach my $pkg ( @Packages ) { >> configure.pl:155: eval "use $pkg"; >> (eval 1)[configure.pl:155]:2: ;>> configure.pl:156: if ( !$@ ) { >> configure.pl:169: if ( $pkg =~ /BackupPC::Lib/ ) { >> configure.pl:184: die <<EOF; BackupPC needs the package version. Please install version before installing BackupPC. Any ideas? Steve |
|
From: Richard S. <hob...@gm...> - 2017-03-22 12:37:33
|
On Tue, Mar 21, 2017 at 11:24 AM, Fritz Elfert <fr...@fr...> wrote: > Is your selinux by any chance enabled? Try setenforce=0 temporarily. If > it works after that, enable it again and after it fails, use audit2allow > to figure out the necessary rulesets. No, pretty sure that's not the case. There is a straightforward error in apache's error_log saying it's not allowing access due to the configuration and when I use a Directory directive to allow access to /usr/sbin it works, but I don't want to allow apache access to everything in /usr/sbin (or /usr/libexec) Thanks, Richard |
|
From: Fritz E. <fr...@fr...> - 2017-03-21 16:23:38
|
Is your selinux by any chance enabled? Try setenforce=0 temporarily. If it works after that, enable it again and after it fails, use audit2allow to figure out the necessary rulesets. CU -Fritz On 21.03.2017 13:50, Richard Shaw wrote: > On Mon, Mar 20, 2017 at 9:36 PM, Kenneth Porter <sh...@se... > <mailto:sh...@se...>> wrote: > > --On Sunday, March 19, 2017 6:41 PM -0500 Richard Shaw > <hob...@gm... <mailto:hob...@gm...>> wrote: > > > The problem is that a compiled binary doesn't belong in /usr/share/... > > (although that's where the current package put it) so I have migrated it > > to /usr/sbin/BackupPC_Admin. > > Would it make more sense to put it in /usr/libexec? > > > I thought about that as well but in either case it doesn't solve the > apache directive issue... > > Thanks, > Richard > > > ------------------------------------------------------------------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > > > > _______________________________________________ > BackupPC-devel mailing list > Bac...@li... > List: https://lists.sourceforge.net/lists/listinfo/backuppc-devel > Wiki: http://backuppc.wiki.sourceforge.net > Project: http://backuppc.sourceforge.net/ > |
|
From: Les M. <les...@gm...> - 2017-03-21 16:19:09
|
On Tue, Mar 21, 2017 at 7:50 AM, Richard Shaw <hob...@gm...> wrote:
>>
>> > The problem is that a compiled binary doesn't belong in /usr/share/...
>> > (although that's where the current package put it) so I have migrated it
>> > to /usr/sbin/BackupPC_Admin.
>>
>> Would it make more sense to put it in /usr/libexec?
>
>
> I thought about that as well but in either case it doesn't solve the apache
> directive issue...
>
Is it possible to use Apache's own suEXEC mechanism?
--
Les Mikesell
les...@gm...
|
|
From: Juergen H. <jue...@un...> - 2017-03-21 16:13:47
|
The Mageia BackupPC 3 package puts BackupPC_Admin (the suid executable) into /var/wwww/backuppc/, next to BackupPC_Admin.cgi - and Apache does not complain. Normally Mageia is configured along the same lines as Fedora. Could this be a viable alternative for BackupPC 4? Juergen |
|
From: Richard S. <hob...@gm...> - 2017-03-21 12:50:09
|
On Mon, Mar 20, 2017 at 9:36 PM, Kenneth Porter <sh...@se...> wrote: > --On Sunday, March 19, 2017 6:41 PM -0500 Richard Shaw > <hob...@gm...> wrote: > > > The problem is that a compiled binary doesn't belong in /usr/share/... > > (although that's where the current package put it) so I have migrated it > > to /usr/sbin/BackupPC_Admin. > > Would it make more sense to put it in /usr/libexec? I thought about that as well but in either case it doesn't solve the apache directive issue... Thanks, Richard |
|
From: Kenneth P. <sh...@se...> - 2017-03-21 02:36:58
|
--On Sunday, March 19, 2017 6:41 PM -0500 Richard Shaw <hob...@gm...> wrote: > The problem is that a compiled binary doesn't belong in /usr/share/... > (although that's where the current package put it) so I have migrated it > to /usr/sbin/BackupPC_Admin. Would it make more sense to put it in /usr/libexec? <https://docs.fedoraproject.org/en-US/Fedora/14/html/Storage_Administration_Guide/s1-filesystem-fhs.html> --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus |
|
From: Johan E. <jo...@mo...> - 2017-03-20 06:27:35
|
Hi all, For the last two years I have been looking at scaling out the backup server end and making backup connections. And since over a decade, I've created solutions for improving backup from behind firewalls and from difficult client machines, while still keeping things mostly automated. At my business, we are currently using BackupPC and UrBackup for different use cases. However, one tool alone could fit the bill with some improvements. Inspired to a large extent by the changes brought by BackupPC version 4, I baked the development ideas and observations from my work into an article. In other words, a big thanks and congratulations goes to Craig and all fellow contributors for reaching version 4! The article can be found here: https://molnix.com/proposal-new-open-source-backup-solution/ I am eager to continue discussions on this topic here! I am also on the UrBackup formus. Maybe some pieces of the proposal end up on a development roadmap somewhere, in some form. Best regards, Johan Ehnberg -- Johan Ehnberg jo...@mo... +358503209688 Molnix Oy molnix.com |
|
From: Richard S. <hob...@gm...> - 2017-03-19 22:41:16
|
>From a Fedora packaging perspective we can't do the standard setup because we don't allow perl scripts to run as suid and can't use mod_perl because it requires a separate apache instance so instead we use a simple C wrapper that can run suid. The problem is that a compiled binary doesn't belong in /usr/share/... (although that's where the current package put it) so I have migrated it to /usr/sbin/BackupPC_Admin. The problem is that I can't seem to find the magic incantation to make apache happy with it. I can duplicate the contents of the Directory directive for /usr/sbin but I don't want to expose that access to everything in /usr/sbin, just BackupPC_Admin, but when I try any version of the Files directive it refuses to work saying something to the effect of: authz_core:error] [pid 32390] [client ::1:37302] AH01630: client denied by server configuration: /usr/sbin/BackupPC_Admin Any ideas? Thanks, Richard |
|
From: Craig B. <cba...@us...> - 2017-03-18 15:46:55
|
Copying reply to users list: I just pushed to git (https://github.com/backuppc/backuppc/commit/ 51e579bae854ec78573cff9c3ff36ee589286502) these changes: - BackupPC exits with status 0 on TERM signal - systemd file now uses Type=simple, dropped the -d, and TERM signal to terminate Craig On Sat, Mar 18, 2017 at 5:36 AM, Richard Shaw <hob...@gm...> wrote: > Since Systemd does not require a service to fork, would it be cleaner to > just call BackupPC without the "-d" option and use "Type=simple"? > > Then we wouldn't even need a pid file... > > Thanks, > Richard > > ------------------------------------------------------------ > ------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > _______________________________________________ > BackupPC-devel mailing list > Bac...@li... > List: https://lists.sourceforge.net/lists/listinfo/backuppc-devel > Wiki: http://backuppc.wiki.sourceforge.net > Project: http://backuppc.sourceforge.net/ > > |
|
From: Richard S. <hob...@gm...> - 2017-03-18 12:36:48
|
Since Systemd does not require a service to fork, would it be cleaner to just call BackupPC without the "-d" option and use "Type=simple"? Then we wouldn't even need a pid file... Thanks, Richard |
|
From: Craig B. <cba...@us...> - 2017-03-16 00:14:40
|
Richard, Unfortunately it would be very difficult for rsync_bpc to support the rsync tests, and there's no plan to do so. Craig On Wed, Mar 15, 2017 at 4:58 PM, Richard Shaw <hob...@gm...> wrote: > While working on submitting rsync-bpc for Fedora I noticed that "make > check" doesn't appear to work due to lots of undefined references. I assume > this has to do with the "bpc_" prefix to a lot of the functions... > > Can this be fixed? It would be nice to know if all tests pass... > > $ make check > gcc -I. -I. -O2 -g -pipe -Wall -Werror=format-security > -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong > --param=ssp-buffer-size=4 -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 > -m64 -mtune=generic -DHAVE_CONFIG_H -Wall -W -c tls.c -o tls.o > gcc -O2 -g -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 > -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 > -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 -m64 > -mtune=generic -DHAVE_CONFIG_H -Wall -W -Wl,-z,relro > -specs=/usr/lib/rpm/redhat/redhat-hardened-ld -o tls tls.o syscall.o > lib/compat.o lib/snprintf.o lib/permstring.o lib/sysxattrs.o -lacl -lpopt > syscall.o: In function `do_chmod': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:114: > undefined reference to `bpc_lchmod' > syscall.o: In function `do_ftruncate': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:153: > undefined reference to `bpc_ftruncate' > syscall.o: In function `do_lutimes': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:242: > undefined reference to `bpc_lutimes' > syscall.o: In function `do_unlink': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:53: > undefined reference to `bpc_unlink' > syscall.o: In function `do_symlink': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:60: > undefined reference to `bpc_symlink' > syscall.o: In function `do_link': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:68: > undefined reference to `bpc_link' > syscall.o: In function `do_lchown': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:79: > undefined reference to `bpc_lchown' > syscall.o: In function `do_mknod': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:87: > undefined reference to `bpc_mknod' > syscall.o: In function `do_rmdir': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:94: > undefined reference to `bpc_rmdir' > syscall.o: In function `do_open': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:104: > undefined reference to `bpc_open' > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:104: > undefined reference to `bpc_open' > syscall.o: In function `do_rename': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:141: > undefined reference to `bpc_rename' > syscall.o: In function `do_mkdir': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:183: > undefined reference to `bpc_mkdir' > syscall.o: In function `do_mkstemp': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:195: > undefined reference to `bpc_mkstemp' > syscall.o: In function `do_stat': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:212: > undefined reference to `bpc_stat' > syscall.o: In function `do_lstat': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:217: > undefined reference to `bpc_lstat' > syscall.o: In function `do_fstat': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:222: > undefined reference to `bpc_fstat' > syscall.o: In function `do_lseek': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:227: > undefined reference to `bpc_lseek' > syscall.o: In function `do_utime': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:264: > undefined reference to `bpc_utime' > lib/sysxattrs.o: In function `sys_lgetxattr': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/lib/sysxattrs.c:31: > undefined reference to `bpc_lgetxattr' > lib/sysxattrs.o: In function `sys_fgetxattr': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/lib/sysxattrs.c:36: > undefined reference to `bpc_fgetxattr' > lib/sysxattrs.o: In function `sys_lsetxattr': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/lib/sysxattrs.c:41: > undefined reference to `bpc_lsetxattr' > lib/sysxattrs.o: In function `sys_lremovexattr': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/lib/sysxattrs.c:46: > undefined reference to `bpc_lremovexattr' > lib/sysxattrs.o: In function `sys_llistxattr': > /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/lib/sysxattrs.c:51: > undefined reference to `bpc_llistxattr' > collect2: error: ld returned 1 exit status > Makefile:101: recipe for target 'tls' failed > make: *** [tls] Error 1 > > Thanks, > Richard > > ------------------------------------------------------------ > ------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > _______________________________________________ > BackupPC-devel mailing list > Bac...@li... > List: https://lists.sourceforge.net/lists/listinfo/backuppc-devel > Wiki: http://backuppc.wiki.sourceforge.net > Project: http://backuppc.sourceforge.net/ > > |
|
From: Richard S. <hob...@gm...> - 2017-03-15 23:58:18
|
While working on submitting rsync-bpc for Fedora I noticed that "make check" doesn't appear to work due to lots of undefined references. I assume this has to do with the "bpc_" prefix to a lot of the functions... Can this be fixed? It would be nice to know if all tests pass... $ make check gcc -I. -I. -O2 -g -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 -m64 -mtune=generic -DHAVE_CONFIG_H -Wall -W -c tls.c -o tls.o gcc -O2 -g -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 -m64 -mtune=generic -DHAVE_CONFIG_H -Wall -W -Wl,-z,relro -specs=/usr/lib/rpm/redhat/redhat-hardened-ld -o tls tls.o syscall.o lib/compat.o lib/snprintf.o lib/permstring.o lib/sysxattrs.o -lacl -lpopt syscall.o: In function `do_chmod': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:114: undefined reference to `bpc_lchmod' syscall.o: In function `do_ftruncate': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:153: undefined reference to `bpc_ftruncate' syscall.o: In function `do_lutimes': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:242: undefined reference to `bpc_lutimes' syscall.o: In function `do_unlink': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:53: undefined reference to `bpc_unlink' syscall.o: In function `do_symlink': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:60: undefined reference to `bpc_symlink' syscall.o: In function `do_link': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:68: undefined reference to `bpc_link' syscall.o: In function `do_lchown': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:79: undefined reference to `bpc_lchown' syscall.o: In function `do_mknod': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:87: undefined reference to `bpc_mknod' syscall.o: In function `do_rmdir': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:94: undefined reference to `bpc_rmdir' syscall.o: In function `do_open': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:104: undefined reference to `bpc_open' /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:104: undefined reference to `bpc_open' syscall.o: In function `do_rename': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:141: undefined reference to `bpc_rename' syscall.o: In function `do_mkdir': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:183: undefined reference to `bpc_mkdir' syscall.o: In function `do_mkstemp': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:195: undefined reference to `bpc_mkstemp' syscall.o: In function `do_stat': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:212: undefined reference to `bpc_stat' syscall.o: In function `do_lstat': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:217: undefined reference to `bpc_lstat' syscall.o: In function `do_fstat': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:222: undefined reference to `bpc_fstat' syscall.o: In function `do_lseek': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:227: undefined reference to `bpc_lseek' syscall.o: In function `do_utime': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/syscall.c:264: undefined reference to `bpc_utime' lib/sysxattrs.o: In function `sys_lgetxattr': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/lib/sysxattrs.c:31: undefined reference to `bpc_lgetxattr' lib/sysxattrs.o: In function `sys_fgetxattr': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/lib/sysxattrs.c:36: undefined reference to `bpc_fgetxattr' lib/sysxattrs.o: In function `sys_lsetxattr': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/lib/sysxattrs.c:41: undefined reference to `bpc_lsetxattr' lib/sysxattrs.o: In function `sys_lremovexattr': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/lib/sysxattrs.c:46: undefined reference to `bpc_lremovexattr' lib/sysxattrs.o: In function `sys_llistxattr': /home/build/rpmbuild/rsync-bpc/BUILD/rsync-bpc-3_0_9_5/lib/sysxattrs.c:51: undefined reference to `bpc_llistxattr' collect2: error: ld returned 1 exit status Makefile:101: recipe for target 'tls' failed make: *** [tls] Error 1 Thanks, Richard |
|
From: Kenneth P. <sh...@se...> - 2017-03-15 06:17:21
|
--On Tuesday, March 14, 2017 5:43 PM +0100 Holger Parplies <wb...@pa...> wrote: > I never seem to be able to remember how many levels of quoting you need, > because I don't use Windoze, but I'm guessing your problem lies there. > What exact settings are you using? I used the web config and trusted that it would quote properly. So my BackupFilesOnly has key "C" and value "/devel" (without quotes). That seems to work fine. --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus |
|
From: Kenneth P. <sh...@se...> - 2017-03-15 06:14:41
|
--On Monday, March 13, 2017 5:38 PM -0700 Craig Barratt <cba...@us...> wrote: > In 4.x you can specify both includes and excludes with rsync and rsyncd. It turns out I misread the v3 documentation. The two are only mutually exclusive with smb backups. So I should be able to tweak some of my configs. --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus |
|
From: Holger P. <wb...@pa...> - 2017-03-14 15:55:25
|
Hi, Craig Barratt wrote on 2017-03-13 16:38:18 -0700 [Re: [BackupPC-devel] RFE: Combing include and exclude in rsyncd backups]: > On Mon, Mar 13, 2017 at 11:08 AM, Kenneth Porter <sh...@se...> > wrote: > > It would be nice if I could combine exclude and include in an rsyncd > > backup. That way I could back up a subdirectory within a module but still > > exclude directories within that directory. Ie. apply the exclude list after > > the include list. Is this a protocol limitation or a BackupPC limitation? > > In 4.x you can specify both includes and excludes with rsync and rsyncd. you can also do that in 3.x. It's actually not really "includes" - BackupFilesOnly is emulated by creating includes and excludes that lead to the desired behaviour. After that, BackupFilesExclude is appended to the list, so you should be able to achieve exactly what you want. For Windoze, I believe the syntax is confusing, because you can sometimes specify / as path separator, sometimes \ and sometimes \\ (quoted backslash). Windoze should accept both slash and backslash for file names, but in-/exclude matching may be just string based on either client or server. I never seem to be able to remember how many levels of quoting you need, because I don't use Windoze, but I'm guessing your problem lies there. What exact settings are you using? Hope that helps. Regards, Holger |
|
From: Holger P. <wb...@pa...> - 2017-03-14 15:40:30
|
Hi,
Alexander Moisseev wrote on 2017-03-13 20:45:49 +0300 [Re: [BackupPC-devel] IPv6 support?]:
> On 3/13/17 7:23 PM, Fritz Elfert wrote:
> > [...]
> > Instead, make a second package "backuppc4". This way, user can have
> > both versions installed and then gracefully migrate.
>
> You can't use both v3 and v4 on the same host as they install files in the
> same place.
that wouldn't need to be true, since distribution packages can choose where to
install which file (and which files possibly to share). However, I can't really
see much point in running both versions on one host aside from brief testing
for migration, and distinct paths for both versions would considerably
complicate migration, somewhat negating the whole point. The distribution would
be left "forever" with somewhat unnatural paths (e.g. /etc/backuppc4,
/var/lib/backuppc4, /usr/share/backuppc4 ...).
I would rephrase your comment as "it shouldn't be possible to install
distribution packages for both v3 and v4 on the same host" ;-).
> However, keeping v3 package "backuppc" and making "backuppc4" package for v4
> is a quite good point as a lot of people use v3 in production and changes
> between v3 and v4 are substantial.
I fully agree with that. Version 4 is a whole new philosophy, with which you
may or may not agree. It's a bit like "upgrading" from init to systemd. No
offence intended ;-).
> BTW v3 support is not dropped by upstream yet.
Another good point. It's a bit awkward to release packages for new v3 releases
that supercede older v3 packages but not older v4 packages if the package name
is identical :-).
Les Mikesell wrote on 2017-03-13 13:14:21 -0500 [Re: [BackupPC-devel] IPv6 support?]:
> My opinion is that nothing should ever break in a 'yum update' from
> CentOS/EPEL repositories if it can possibly be avoided. And even
> with v4 in release status there may be reasons to install v3 or
> rebuild a machine without changing anything.
I also agree with that (and would extend it to other distros).
> Using different package names will let them co-exist for at least as long
> as v3 will be maintained and let the sysadmin decide when to switch.
> I might think differently if the upgrade could be completely transparent,
> though.
I don't see how a transparent upgrade would make the reasons go away, for
which you might want to install v3 or decide when to switch. (*)
While running v3 and v4 on the same host does not sound like a good idea,
running them on *different* hosts definitely does!
Regards,
Holger
(*) Aside from that, it is not *possible* to guarantee a transparent upgrade.
My RsyncClientCmd might invoke a remote end that is dependent on the exact
sequence of arguments passed by File::RsyncP, so upgrading might even
involve changing and recompiling code on a remote machine. The normal cases
(ssh, sudo, ssh + sudo) might work, but the flexibility the configuration
mechanism gives you comes at the cost of not being able to predict what
people will actually do.
|
|
From: Craig B. <cba...@us...> - 2017-03-13 23:46:44
|
The configure.pl script updates the relevant rsync settings in the main
config.pl file. In particular, it updates $Conf{RsyncArgs}
and $Conf{RsyncRestoreArgs} with the 4.0 settings.
$Conf{RsyncSshArgs} is a new 4.0 setting, and $Conf{RsyncClientCmd}
and $Conf{RsyncClientRestoreCmd} are no longer used. configure.pl tries to
extract a sensible default setting for $Conf{RsyncSshArgs} from
$Conf{RsyncClientCmd}.
The potential problems are that there are per-client overrides for these
settings (which is not checked by configure.pl), and the computed value for
$Conf{RsyncSshArgs} isn't correct.
All the major per-client settings (schedule, what to backup etc) are
compatible from 3.x to 4.x.
Craig
On Mon, Mar 13, 2017 at 11:09 AM, Kenneth Porter <sh...@se...>
wrote:
> --On Monday, March 13, 2017 11:52 AM -0700 Craig Barratt
> <cba...@us...> wrote:
>
> > That said, the 4.0 client configuration has changed for rsync hosts due
> to
> > the use of rsync_bpc, so a package upgrade from 3.x to 4.x will
> > potentially require some admin effort to update the client configs. So I
> > agree it won't be a completely seamless upgrade.
>
> Can you say more about what needs to change? I don't think I'm doing
> anything incredibly exotic. Mainly a few hosts use include instead of
> exclude to allow me to split up backups into smaller units without changing
> the rsyncd config on the PC being backed up.
>
> ---
> This email has been checked for viruses by Avast antivirus software.
> https://www.avast.com/antivirus
>
>
> ------------------------------------------------------------
> ------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> BackupPC-devel mailing list
> Bac...@li...
> List: https://lists.sourceforge.net/lists/listinfo/backuppc-devel
> Wiki: http://backuppc.wiki.sourceforge.net
> Project: http://backuppc.sourceforge.net/
>
|
|
From: Craig B. <cba...@us...> - 2017-03-13 23:38:45
|
In 4.x you can specify both includes and excludes with rsync and rsyncd. Craig On Mon, Mar 13, 2017 at 11:08 AM, Kenneth Porter <sh...@se...> wrote: > It would be nice if I could combine exclude and include in an rsyncd > backup. That way I could back up a subdirectory within a module but still > exclude directories within that directory. Ie. apply the exclude list after > the include list. Is this a protocol limitation or a BackupPC limitation? > > Example: > > Host provides "C" module exposing C:\ for backup. I have backups like this: > > C\devel\Boost (just the Boost C++ libraries) > C\devel (all development except Boost) > C (the rest of the drive) > > Currently I have to list all the directories under C\devel for the include > list. I'd rather have it recognize an exclude list of just C\devel\Boost so > that when a user adds a new directory under C\devel I don't have to modify > the backup configuration. > > > > --- > This email has been checked for viruses by Avast antivirus software. > https://www.avast.com/antivirus > > > ------------------------------------------------------------ > ------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > _______________________________________________ > BackupPC-devel mailing list > Bac...@li... > List: https://lists.sourceforge.net/lists/listinfo/backuppc-devel > Wiki: http://backuppc.wiki.sourceforge.net > Project: http://backuppc.sourceforge.net/ > |
|
From: Les M. <les...@gm...> - 2017-03-13 18:14:28
|
On Mon, Mar 13, 2017 at 12:52 PM, Craig Barratt
<cba...@us...> wrote:
> BackupPC 4.0 is backward compatible with 3.x backup storage - they can be
> viewed/restored etc. I just pushed an optional migration tool,
> BackupPC_migrateV3toV4, that replaces the hardlinked 3.x storage in each
> backup with 4.x-style attrib files and reference counting. So you can use
> that to get rid of all the 3.x hardlinks if you want, but it's not
> necessary.
>
> That said, the 4.0 client configuration has changed for rsync hosts due to
> the use of rsync_bpc, so a package upgrade from 3.x to 4.x will potentially
> require some admin effort to update the client configs. So I agree it won't
> be a completely seamless upgrade.
>
> What do other people recommend in terms of having a new package name for
> BackupPC 4.x?
>
My opinion is that nothing should ever break in a 'yum update' from
CentOS/EPEL repositories if it can possibly be avoided. And even
with v4 in release status there may be reasons to install v3 or
rebuild a machine without changing anything. Using different package
names will let them co-exist for at least as long as v3 will be
maintained and let the sysadmin decide when to switch. I might think
differently if the upgrade could be completely transparent, though.
--
Les Mikesell
les...@gm...
|
|
From: Kenneth P. <sh...@se...> - 2017-03-13 18:09:53
|
--On Monday, March 13, 2017 11:52 AM -0700 Craig Barratt <cba...@us...> wrote: > That said, the 4.0 client configuration has changed for rsync hosts due to > the use of rsync_bpc, so a package upgrade from 3.x to 4.x will > potentially require some admin effort to update the client configs. So I > agree it won't be a completely seamless upgrade. Can you say more about what needs to change? I don't think I'm doing anything incredibly exotic. Mainly a few hosts use include instead of exclude to allow me to split up backups into smaller units without changing the rsyncd config on the PC being backed up. --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus |