- 30 Jun, 2022 1 commit
-
-
Eric Bollengier authored
-
- 25 Mar, 2022 1 commit
-
-
Eric Bollengier authored
-
- 24 Mar, 2022 5 commits
-
-
Eric Bollengier authored
-
Eric Bollengier authored
-
-
This feature can be used -only- if all of the daemons (DIR, SD and FD) are updated to this version. When the new 'o' accurate option is used, only metadata is being backed up in case file's contents did not change (meaning that file's checksum stays the same). New stream was added to handle that during the backup and restore phase.
-
Eric Bollengier authored
-
- 29 Apr, 2021 1 commit
-
-
This commit is the result of the squash of the following main commits: Author: Eric Bollengier <eric@baculasystems.com> Date: Thu Dec 19 11:07:42 2019 +0100 win32: Fix #5801 about incorrect File=/ behavior with subvolumes in backup and estimate jobs Author: Eric Bollengier <eric@baculasystems.com> Date: Fri Dec 14 17:05:26 2018 +0100 Fix memory issue with the strip fileset option Author: Alain Spineux <alain@baculasystems.com> Date: Mon Nov 12 11:07:04 2018 +0100 win32: Fix #4286 about the restore of a volume root directory with the hidden & system attributes - Detect root directories like C:\ or C:\mount_point and setup st_rdev=WIN32_ROOT_POINT to tell restore to not restore the hidden and system attributes - Be aware that there is 2 different "directories" (1) the parent directory like c:\mount_point and (2) the root directory of the volume that your are attaching to the parent. The attributes that you see is the attribute of (1) but for drive letters, their is no (1) then you see always (2) (that are hidden and system). When VSS snapshot is used Author: Alain Spineux <alain@baculasystems.com> Date: Tue Aug 28 17:14:31 2018 +0200 snapshot: maintain both paths instead of stripping and unstripping - snap_fname is the path inside the snapshot used for "system" operation - fname is the path from the FileSet used for display and for SD or DIR "computation" - when snapshot is not in used, both are equal Author: Alain Spineux <alain@baculasystems.com> Date: Sat Apr 28 07:42:58 2018 +0200 win32: Rewrite of the stat() and reparse point detection - combine all stat() functions - statDir() i& stat2() are not used anymore - adjuste find_one for directory symlink
-
- 22 Jun, 2020 1 commit
-
-
Eric Bollengier authored
-
- 11 Aug, 2018 1 commit
-
-
Kern Sibbald authored
-
- 01 Jul, 2018 1 commit
-
-
- when a path > 250, some snapshot related POOLMEM inside FF_PKT are enlarged, but the FF_PKT copy done before to jump into a dir stil use the old POOLMEM
-
- 20 Apr, 2018 1 commit
-
-
Kern Sibbald authored
-
- 04 Dec, 2017 1 commit
-
-
- breaddir() is a thread safe replacement for readdir_r() - remove any usage of readdir_r() and readdir() in the core - remove NAMELEN() (that was never used except for 3 uses in find_one.c)
-
- 01 Nov, 2017 1 commit
-
-
Kern Sibbald authored
-
- 21 Dec, 2015 1 commit
-
-
Kern Sibbald authored
-
- 20 Dec, 2015 1 commit
-
-
Kern Sibbald authored
-
- 13 Aug, 2015 1 commit
-
-
- 27 Mar, 2014 1 commit
-
-
Kern Sibbald authored
-
- 20 Apr, 2013 6 commits
-
-
-
Kern Sibbald authored
-
-
-
-
-
- 05 Nov, 2010 1 commit
-
-
ricozz authored
-
- 02 Aug, 2010 2 commits
-
-
- 18 Sep, 2009 1 commit
-
-
From one side, access(2) is not reliable by definition. According to specification, access(2) uses process's real UID and GID instead of effective IDs, and it ignores process capabilities. Consequently, privileged process with unprivileged real IDs cannot use access(2) to check its access to files. From another side, the access(2) directory check looks redundant because the directory is opened right below using opendir(3).
-
- 17 Sep, 2009 1 commit
-
-
Kern Sibbald authored
-
- 09 Sep, 2009 1 commit
-
-
ricozz authored
-
- 03 Aug, 2009 1 commit
-
-
ricozz authored
-
- 15 Mar, 2009 1 commit
-
-
Kern Sibbald authored
restore two databases (only one is permitted). This responds to bug #1234. kes Apply the nodump patch supplied by Frank Kardel that fixes the NODUMP flag problem. This fixes bug #1221 git-svn-id: https://bacula.svn.sourceforge.net/svnroot/bacula/trunk@8536 91ce42f0-d328-0410-95d8-f526ca767f89
-
- 13 Oct, 2008 1 commit
-
-
Kern Sibbald authored
kes Implement code to prohibit ExcludeDirContaining in an Exclude section (same for Plugin and Options). A bit of a kludge with RES_ITEM2 ... git-svn-id: https://bacula.svn.sourceforge.net/svnroot/bacula/trunk@7775 91ce42f0-d328-0410-95d8-f526ca767f89
-
- 14 Jul, 2008 1 commit
-
-
Kern Sibbald authored
kes First non-working cut of vbackup kes Correct FD heartbeat code to use volatile variable accessed by two threads. kes Zero thread id with memset if on Win32 in case it is a struct. kes If cannot create JobMedia record, continue to try to write eof and clean up in block.c git-svn-id: https://bacula.svn.sourceforge.net/svnroot/bacula/trunk@7380 91ce42f0-d328-0410-95d8-f526ca767f89
-
- 06 Jul, 2008 1 commit
-
-
Kern Sibbald authored
kes Clear hash packet for hardlinked files correctly. This corrects a long standing bug where hardlinked files selected individually would occasionally not be restored. git-svn-id: https://bacula.svn.sourceforge.net/svnroot/bacula/trunk@7324 91ce42f0-d328-0410-95d8-f526ca767f89
-
- 21 May, 2008 1 commit
-
-
Kern Sibbald authored
many were suppressed. This should fix bug #1070. kes Fix Win32 reparse points. Bacula will not recurse into any reparse point directory, including mount points, unless the directory is explicitly mentioned at the top level (same as with Unix). A file that is linked to another file will be backed up -- much as Unix does for hardlinked files. This fixes bug #1041. git-svn-id: https://bacula.svn.sourceforge.net/svnroot/bacula/trunk@7004 91ce42f0-d328-0410-95d8-f526ca767f89
-
- 02 May, 2008 1 commit
-
-
Kern Sibbald authored
Bacula trunk. Regress now works with libdbi. Nice. kes Apply .nobackup patch from bug #1077 submitted by Edwin Groothuis. kes Apply Bastian Friedrich's patch that adds %p to edit the pool name into jobs cloned via the run directive. git-svn-id: https://bacula.svn.sourceforge.net/svnroot/bacula/trunk@6874 91ce42f0-d328-0410-95d8-f526ca767f89
-
- 01 May, 2008 1 commit
-
-
Kern Sibbald authored
This fixes bug #1078. kes Remove 50 millisec wait in SD that broke debugger. git-svn-id: https://bacula.svn.sourceforge.net/svnroot/bacula/trunk@6864 91ce42f0-d328-0410-95d8-f526ca767f89
-
- 25 Feb, 2008 1 commit
-
-
ricozz authored
git-svn-id: https://bacula.svn.sourceforge.net/svnroot/bacula/trunk@6486 91ce42f0-d328-0410-95d8-f526ca767f89
-
- 09 Feb, 2008 1 commit
-
-
Kern Sibbald authored
find_one_file. git-svn-id: https://bacula.svn.sourceforge.net/svnroot/bacula/trunk@6385 91ce42f0-d328-0410-95d8-f526ca767f89
-