Axel Burri
9ae823a757
btrbk: cosmetics: remove too chatty debug log
2018-05-10 15:44:46 +02:00
Vít Novotný
c8fcb01bc2
btrbk: bugfix: do not read target btrfs tree on "btrbk snapshot --preserve"
...
If running "btrbk snapshot --preserve", there is no need to initialize
targets, and we don't want to warn and fail (exitcode=10) on missing
targets.
2018-05-10 15:18:36 +02:00
Axel Burri
17f41118d3
btrbk: bugfix: compare all mountpoints while resolving
...
When comparing longest match in mounts, all mountpoints have to be
taken into account (not only the btrfs mount points).
2018-05-10 14:02:44 +02:00
Axel Burri
1c83a6545d
btrbk: add filter capabilities to vinfo_subvol_list
...
Clean up (and speedup) code by filtering while building the list
instead of checking every item after building.
2018-05-10 11:48:34 +02:00
Axel Burri
a25487e873
btrbk: cosmetics (log messages)
2018-05-10 11:48:34 +02:00
Axel Burri
ef5c369a37
btrbk: use _is_same_fs_tree() where applicable
2018-05-10 11:48:34 +02:00
Axel Burri
0454f60ad1
btrbk: bugfix: match btrbk_basename in get_latest_snapshot_child()
2018-05-10 11:48:34 +02:00
Axel Burri
2c1c3b4d54
btrbk: cleanup: remove snapshot_dir, rename sroot->snaproot
2018-05-10 11:48:34 +02:00
Axel Burri
c457540ce3
btrbk: use separate vinfo for snapshot directory (allows snapshot_dir to be a mountpoint)
...
Instead of passing snapshot_dir all over the place, use a separate
vinfo for the snapshot directory, accessible by vinfo_snapshot_root().
As it is initialized separately by vinfo_init_root(), it can be on a
different mountpoint.
This also allows us to use different semantics for snapshot_dir in the
future, as it does not need to be relative to the volume directory.
2018-05-10 11:48:34 +02:00
Axel Burri
f5dc4e0a36
btrbk: add known mountpoints to btr_tree nodes as anchor for reverse lookup
2018-05-10 11:48:34 +02:00
Axel Burri
e9374b3b1d
btrbk: replace url_cache by spec_cache
2018-05-10 11:48:34 +02:00
Axel Burri
0ea0430c43
btrbk: cleanup (cosmetics, documentation)
2018-05-10 11:48:34 +02:00
Axel Burri
b37ef84e36
btrbk: always read mountpoints; include all snapshots from mountpoint as candidates for best common parent
...
Dropped readin of subvolid and realpath by btrfs_subvolume_show(), we
now always read /proc/self/mounts (and call readlink).
When picking the best common parent in get_best_parent(), we want to
list as many snapshots as possible. For now, we list all from the
mountpoint of snaproot ($sroot/<snapshot_dir>), due to a bug in
btrfs-progs [1]. Also added code (commented out) to list snapshots
from all known mountpoints.
[1] https://github.com/kdave/btrfs-progs/issues/96
2018-05-10 11:48:05 +02:00
Axel Burri
b549e11b43
btrbk: raw targets: move tree readin to separate function; add caching
...
Adds new function vinfo_init_raw_root(), similar to vinfo_init_root().
Note that we don't (yet) resolve realpath for raw targets.
2018-05-10 11:36:21 +02:00
Axel Burri
7a1bc257c1
btrbk: raw targets: create fake btr_tree instead of maintaining separate list
2018-05-10 11:36:21 +02:00
Axel Burri
6c502cbdcc
btrbk: search complete target tree for correlated subvolumes
...
- move matching for correlated subvolumes from get_receive_targets
into new function _receive_target_nodes
- add lookup tables in btr_tree (RECEIVED_UUID_HASH, UUID_HASH),
allowing for faster matching in _receive_target_nodes
- add vinfo_resolved() for mapping nodes to vinfo
- rename get_latest_common to get_best_parent (while moving some
functionality to new function get_related)
- cleanup
2018-05-10 11:36:21 +02:00
Axel Burri
490f680f41
btrbk: bugfix: add backend mappings for "btrfs filesystem usage"
2018-04-30 14:15:30 +02:00
Axel Burri
719fb5fb74
btrbk: bugfix: correct scheduling of "first weekly backup in month/year"
...
In the scheduler, a month (or year) does not start at the first day,
but at the first `preserve_day_of_week`. Make sure that all days
before `preserve_day_of_week` in a month get delta_months+1.
Example (corner case):
- `preserve_day_of_week sunday`
- `target_preserve *m`
- no backups in 2018-02
- backup with timestamp 2018-03-01 (which is a thursday)
- backup with timestamp 2018-03-04 (which is a sunday)
Without this patch, because there are no sunday backups in 2018-02,
the first backup is considered a weekly (+4d after sunday), and as
such "first weekly of month 2018-03", and the second one is discarded.
With this patch, the first item is considered "first weekly of month
2018-02", and the second gets "first weekly of month 2018-03".
NOTE: This change may result in (previously preserved) backups to be
deleted!
2018-04-05 18:00:29 +02:00
Axel Burri
e4a0938712
btrbk: ignore "preserve_hour_of_day" on snapshots/backups having "timestamp_format=short"
...
Snapshots and backups having no exact time information (created with
"timestamp_format=short") are set to 00:00, which would be regarded as
"previous day" if preserve_hour_of_day is greater than 0. Fix this by
ignoring preserve_hour_of_day in this case.
2018-04-05 18:00:29 +02:00
Axel Burri
c89ffd4213
btrbk: schedule: format preserve text only if needed
...
No functional changes, improves performance of schedule().
2018-04-05 18:00:24 +02:00
Thiodwitnir
5791d72171
btrbk: add "preserve_hour_of_day" configuration option
...
Introduces the new config option "preserve_hour_of_day" to specify
after what time backups should be considered as dailies.
Based on pull request #204 , with changes:
- calculation of weekly backups
- change format of preserve_matrix
2018-04-05 16:37:31 +02:00
Axel Burri
951ae256fa
btrbk: bump version to 0.27.0-dev
2018-04-05 16:26:57 +02:00
Axel Burri
1fd3a547df
change version to 0.26.1; bump copyright year
2018-03-05 11:55:56 +01:00
Axel Burri
6f38536052
btrbk: set c_default for config keys with computed default
...
Suppress "Option redefined" warning for snapshot_name config option,
which has hardcoded (computed) default already set when checking.
fix regression: 0ebe2ea2e1
2018-02-26 16:06:12 +01:00
Axel Burri
e75cf0c72c
btrbk: fix exit status and transaction log when archive_exclude is set
...
Similar to ABORTED=USER_SKIP (active commandline filter), archives
having ABORTED=ARCHIVE_EXCLUDE_SKIP (active archive_exclude
configuration) do not cause exit status 10 and are hidden from
transaction log.
2018-02-26 14:35:26 +01:00
Axel Burri
5651c96e4e
btrbk: archive: always print "[-]" if no action was performed for target
2018-02-14 00:18:44 +01:00
Axel Burri
8610e75459
btrbk: add archive_exclude configuration option
...
Support wildcard characters, matches against both "$sroot->{PATH}" and
"$sroot->{PATH}/$snapshot_name".
2018-02-13 22:50:23 +01:00
Axel Burri
0ebe2ea2e1
btrbk: add "allow_multiple" flag for config_options declaration; use for option "group"
2018-02-13 19:30:24 +01:00
Axel Burri
3aaafa3d88
btrbk: add "meta" context for config defaults
...
Preparatory patch to allow multiple identical config keys in same
section.
2018-02-13 19:28:38 +01:00
Axel Burri
b2b43cf199
btrbk: add configurable error_statement for check functions
...
Corrects error messages from --exclude option.
Applies to in check_file(), check_url() and append_config_option().
2018-02-13 17:40:14 +01:00
Axel Burri
db1fe2d11a
btrbk: fix output of "btrbk list backups" and "btrbk stats" (match all uuid/received_uuid combinations)
2018-02-07 16:24:49 +01:00
Axel Burri
a3641cff74
btrbk: fix filter statement matching for volume="/"
...
While $vol->{URL} can contain "//" if volume="/" (intentionally, this
is an assembled path), the filter statements are sanitized using
check_url(). This means we need to match the filter statement against
check_url($vol->{URL}). Same applies to subvol.
2018-02-03 12:55:21 +01:00
Axel Burri
90fed6525e
btrbk: bugfix: restrict snapshot_dir and snapshot_create to root/volume/subvolume context
2018-01-31 19:01:00 +01:00
Axel Burri
753d68b2ed
btrbk: add documentation on caching in btr_tree()
2018-01-30 14:02:36 +01:00
Axel Burri
84bac59ab4
btrbk: fix parsing of "openssl_iv_size" configuration option
2018-01-21 18:53:29 +01:00
Axel Burri
13f37491ed
btrbk: cosmetics: change order of commands in help message
2017-10-12 17:08:46 +02:00
Axel Burri
79cbf7c0c3
change version to 0.26.0
2017-10-12 00:39:06 +02:00
Axel Burri
909c68f164
btrbk: transaction_log and transaction_syslog are only allowed in root context
2017-10-11 20:54:41 +02:00
Axel Burri
d941ecfce9
documentation: change digint links to https://
2017-10-11 20:54:41 +02:00
Axel Burri
e3ee674085
btrbk: limit search depth to 256 when resolving ancestors
...
Note that the current implementation is not very optimized, we should
introduce a parent/child hash table for all queries in
get_latest_common().
2017-10-10 16:46:05 +02:00
Axel Burri
3be65b9f67
btrbk: resolve ancestors (recursive on parent_uuid chain) when searching for latest common subvolume
2017-10-10 13:10:33 +02:00
Axel Burri
0799820556
btrbk: support btrfs-progs 4.13.2: adapt parsing of "btrfs subvolume list"
...
btrfs-progs v4.13.2 changed the output of "btrfs subvolume list", in
order to make the tabular output format look nicer (?!?):
e4c6772f69
2017-10-09 23:04:07 +02:00
Axel Burri
e9a517f161
btrbk: add options "{snapshot,target,archive}_qgroup_destroy": destroy qgroups whenever a subvolume is deleted
...
Btrfs does not destroy qgroups when subvolumes are deleted (see
https://bugzilla.kernel.org/show_bug.cgi?id=91751 ). As a workaround
for this, btrbk can be configured to always destroy the corresponding
default qgroup "0/<subvol-id>" whenever a subvolume (snapshot, backup
or archive) is deleted.
Added configuration options:
- snapshot_qgroup_destroy
- target_qgroup_destroy
- archive_qgroup_destroy
2017-10-02 16:23:23 +02:00
Axel Burri
6cb98700d8
btrbk: allow archive_preserve_* options in root context only
2017-10-02 13:54:37 +02:00
Axel Burri
2e974c1f4f
btrbk: add "--wipe" command line option
2017-09-28 14:17:00 +02:00
Axel Burri
5bdc3e527b
btrbk: add "prune" command
2017-09-28 14:17:00 +02:00
Axel Burri
2809dc54c5
btrbk: remove sorting of deleted subvolumes in summary (not needed anymore)
...
Deleted subvolumes are already sorted by schedule().
Sorting by PATH is bad anyways, as it does not sort _NN postfix
correctly.
2017-09-28 14:17:00 +02:00
Axel Burri
512aca5332
btrbk: parse output of "btrfs subvolume delete"
...
When doing a batch delete (multiple deletes with one call to "btrfs
subvolume delete"), we want to know which subvolumes have failed. For
this, we need parse the error output.
On any parsing failure, we assume that nothing has been deleted, and
warn accordingly (forward compatibility).
2017-09-28 14:17:00 +02:00
Axel Burri
9d9527ca9a
btrbk: simplify transaction function calls
...
Prefix transaction status with "dryrun_" in start_transaction() /
end_transaction if $dryrun is set.
2017-09-28 14:17:00 +02:00
Axel Burri
422d52c063
btrbk: add key derivation for encrypted raw targets using external backend
2017-09-28 14:17:00 +02:00
Axel Burri
de7628ac7c
btrbk: add openssl_enc encryption for raw targets; add system_urandom()
...
Example:
Manually create a key:
# KEYFILE=/some/secure/place/btrbk.key
# dd if=/dev/urandom bs=1 count=32 | od -x -A n | tr -d "[:space:]" > $KEYFILE
btrbk.conf:
volume /mnt/btr_pool
incremental no
raw_target_encrypt openssl_enc
openssl_ciphername aes-256-cbc
openssl_iv_size 16 # NOTE: set to "no" if no IV is needed by the selected cipher
openssl_keyfile /some/secure/place/btrbk.key
subvolume home
target raw ssh://cloud.example.com/backup
2017-09-28 14:17:00 +02:00
Axel Burri
251c2fb2a1
btrbk: re-enable parsing of deprecated raw file format (uuid suffix)
2017-09-28 14:17:00 +02:00
Axel Burri
e804930b5e
btrbk: allow deletion of non-incremental raw targets
...
implemented directly in btrfs_subvolume_delete()
2017-09-28 14:17:00 +02:00
Axel Burri
cd8d7e3a0a
btrbk: use sidecar file "*.info" instead of encoding uuids into filename for raw targets
...
pros:
- better forward compatibility, e.g. symmetrical encryption
- better readability of files
cons:
- two files per backup
2017-09-28 14:16:53 +02:00
Axel Burri
571dae4428
btrbk: do not run in perl taint mode by default: remove "perl -T" in hashbang; hardcode $PATH only if taint mode is enabled
...
While taint mode [1] is a nice feature of perl, e.g. it disallows
using variables (such as filenames from the config file) which were
not validated in system() commands, it also treats $PATH as insecure
(which inherently is, as perl cannot know who messed around with it).
[1] perlsec(1): http://perldoc.perl.org/perlsec.html
[2] perlrun(1): http://perldoc.perl.org/perlrun.html
Note that btrbk still does all taint checks, and can be run in taint
mode:
- by executing `perl -T /usr/sbin/btrbk`,
- or by changing the hashbang to: `!#/usr/bin/perl -T`.
2017-09-25 17:53:35 +02:00
Axel Burri
5f867c2347
btrbk: run_cmd(): catch all possible errors when executing system command
2017-09-25 15:00:40 +02:00
Axel Burri
e402435dc8
btrbk: treat deprecated warnings as non-fatal
...
Despite FATAL warnings are discouraged in perl and may break forward
compatibility [1], we still use it as btrbk is usually run as root and
we really want perl to die on programmatic errors.
[1] "perldoc warnings"
2017-09-25 15:00:40 +02:00
Axel Burri
6c2d56d621
btrbk: action "clean": dont list failed deletes in summary
2017-09-21 12:39:48 +02:00
Axel Burri
f113436626
btrbk: bugfix: check path when expanding wildcards (warn and ignore if check fails)
...
Before this patch, btrbk errored (unsafe command), and then died.
Fixes #181 .
2017-09-11 18:49:14 +02:00
boerwastaken
d265c34149
btrbk: make SSH minimally chatty (always add '-q' option)
...
Fixes digint/btrbk#179 : When SSH is chatty (informing of host key
rollover etc) the remote command output parser borks.
2017-09-04 13:00:19 +02:00
Axel Burri
8eb88a8681
documentation: replace POD header with comments
2017-08-28 20:33:00 +02:00
Axel Burri
88aa8c1fea
btrbk: remove "duration" column from transaction_log/transaction_syslog
...
The "duration" column in the transaction log has proven to be
confusing to some users, especially on errors (e.g. "send-receive
ERROR 27" in issue #177 ). As it's not really necessary (duration can
be computed from the corresponding "starting" log entry), it's now
being dropped.
2017-08-28 17:55:27 +02:00
Axel Burri
6acea6a08d
btrbk: bugfix: print correct time in end_transaction()
2017-08-28 17:54:17 +02:00
Axel Burri
5428e9cd93
btrbk: add "resume" and "snapshot" commands; add --preserve-snapshots and --preserve-backups options; deprecate "-r, --resume-only" command line option
2017-08-21 13:23:20 +02:00
Axel Burri
a9f1b6b24a
change version to 0.25.1
2017-07-30 16:03:58 +02:00
Axel Burri
6cf5d59644
btrbk: add support for btrfs-progs 4.12: do not fail or set realpath_cache for relative paths in btrfs_subvolume_show()
...
As of btrfs-progs-v4.12, the "btrfs subvolume show" command does not
print the full (absolute, resolved) path anymore [1]. Instead, it prints
the relative path to btrfs root (or "/" if it is the root).
The impact for btrbk is that we cannot fill our realpath_cache in
btrfs_subvolume_show() anymore. This is not fatal, but has the
following consequences:
- The "check for duplicate snapshot locations" may now miss
subvolumes specified by symlinks.
- If multiple "volume" sections point to the same subvolume (e.g. if
specified using symlinks) an additional "btrfs subvolume list" is
called. Note that the subvolume will still be recognized as
identical, and the btr_tree will not be rebuilt.
[1] btrfs-progs commit: b7df24aa5cddc4802b9938f56372b73869775cd9
2017-07-30 15:25:32 +02:00
Axel Burri
26682213e2
btrbk: cleanup btrfs_subvolume_list() and comments
2017-07-30 15:03:23 +02:00
Axel Burri
356231bc2a
btrbk: abort if source subvolume has readonly or received_uuid flag
2017-07-29 19:03:23 +02:00
Axel Burri
c0c546eb4e
btrbk: stats command: show "correlated" status instead of "orphaned" (inverse logic which caused a lot of confusion)
2017-06-25 21:16:08 +02:00
Axel Burri
15bae5e051
btrbk: stats command: print up-to-date status on backups
2017-06-25 21:14:03 +02:00
Axel Burri
b2cc99b0fb
btrbk: bugfix: accept "no" for "transaction_log", "transaction_syslog" and "lockfile" configuration options
2017-06-25 18:06:06 +02:00
Axel Burri
315b3f24a1
btrbk: add config option "stream_buffer <size>". if set, pipes send stream through "mbuffer -m <size>"
2017-06-08 12:35:22 +02:00
Axel Burri
7436d9432c
change version to 0.25.0
2017-03-18 16:53:07 +01:00
Axel Burri
d31ffb8cb0
btrbk: allow trailing comments in btrbk.conf
2017-03-18 16:34:53 +01:00
Axel Burri
358a2b1169
btrbk: cleanup: btrfs_send_to_file() takes no more additional options (all info is fetched directly from source/target config)
2017-03-18 15:06:48 +01:00
Axel Burri
28c65e4675
btrbk: add raw_target_split option, using "split" instead of "dd" to write raw files
2017-03-18 14:47:43 +01:00
Axel Burri
3b7ede773a
btrbk: bugfix: make sure rate limiting comes after compression in cmd_pipe
2017-03-18 12:18:47 +01:00
Axel Burri
cb82bd5fa4
btrbk: add supoort for "backend btrfs-progs-sudo" option; adaptions in ssh_filter_btrbk.sh
2017-01-04 15:04:22 +01:00
Ian Kelling
d02f67a924
btrbk: get_receive_targets: also match "src.received_uuid == target.uuid"
...
This gets important when using an old backup disk as source.
In terms of btrfs send/receive, all subvolumes matching "uuid /
received_uuid" are valid backups.
Merged (amend) from pull request: #116
Verified by Axel Burri <axel@tty0.ch>
2017-01-04 12:41:05 +01:00
Axel Burri
79637de5aa
btrbk: raw_target_encrypt: always set "gpg --no-random-seed-file": prevents creation of "~/.gnupg/random_seed" with slight perfomance penalty.
...
We set "--no-random-seed-file" because one of the btrbk
design principles is to not create any files unasked. Enabling
"--no-random-seed-file" creates ~/.gnupg/random_seed, and as
such depends on $HOME to be set correctly (think on running in
cron). From gpg2(1) man page:
--no-random-seed-file GnuPG uses a file to store its
internal random pool over invocations This makes random
generation faster; however sometimes write operations are not
desired. This option can be used to achieve that with the cost
of slower random generation.
2017-01-03 13:41:39 +01:00
Axel Burri
b69e9ebf34
btrbk: show aggregate "size" and "used" for "usage" action
2016-12-11 15:36:48 +01:00
Axel Burri
da849b37ae
change version to 0.24.0
2016-11-16 15:32:17 +01:00
Axel Burri
ec63e9932b
btrbk: support for btrfs-progs v4.8.3
...
Fix parsing of "btrfs sub show" output, which has changed for toplevel
subvolume.
2016-11-16 15:02:49 +01:00
Axel Burri
57be688f6b
btrbk: bugfix: fix "list latest" with no snapshots
2016-11-12 13:21:33 +01:00
Axel Burri
a8a311c8c7
btrbk: add "backend" configuration option: support for distinct binaries from btrfs-progs-btrbk
...
btrfs-progs-btrbk sources (fork of btrfs-progs) can be found at:
https://github.com/digint/btrfs-progs-btrbk
2016-08-29 18:40:42 +02:00
Axel Burri
dc5dee3121
btrbk: use loglevel trace (instead of debug) for repeated WARN_ONCE
2016-08-29 15:03:56 +02:00
Axel Burri
1aa208151b
btrbk: improve check_file(): sanitize only on demand
2016-08-29 15:03:20 +02:00
Axel Burri
4c4afe776d
btrbk: skip target metadata test if send/receive has errors
2016-08-29 15:03:11 +02:00
Axel Burri
961f96833a
btrbk: use "dd" for raw target output, with configurable block size (raw_target_block_size)
...
We use "dd" instead of shell redirections, as it is common to have
special filesystems (like NFS, SMB, FUSE) mounted on the raw target
path. By using "dd" we make sure to write in reasonably large blocks
(default=128K), which is not always the case when using redirections
(e.g. "gpg > outfile" writes in 8K blocks).
Another approach would be to always pipe through "cat", which uses
st_blksize from fstat(2) (with a minimum of 128K) to determine the
block size.
2016-08-25 13:14:01 +02:00
Axel Burri
3dabb507e6
btrbk: do not set "--compress-algo none" on gpg command if raw_target_compress=no
...
Note that gpg might still compress (according to OpenPGP standard) if
--compress-algo is not set. This is generally a good thing.
2016-08-25 12:53:12 +02:00
Axel Burri
b0feaf4413
btrbk: force disabling of ssh compression (ssh -o compression=no) if stream_compress is set
2016-08-25 12:53:12 +02:00
Axel Burri
b49ee61ecd
btrbk: run_cmd: cosmetics
2016-08-25 12:53:12 +02:00
Axel Burri
5f530e47e5
btrbk: run_cmd: only perform sanity checks on files (hardcoded parts are considered safe)
2016-08-25 12:53:06 +02:00
Axel Burri
96f0d4c175
btrbk: fix and cleanup stream compression
2016-08-24 15:25:48 +02:00
Axel Burri
13d27c8616
btrbk: perform extra metadata check on target subvolume after "btrfs receive"
...
Eliminates error cases where "btrfs receive" input is null, at the
cost of an additional call to "btrfs subvolume show".
2016-08-19 17:41:03 +02:00
Axel Burri
deeb12c069
ssh_filter_btrbk: allow stream compression if --compress option is set.
2016-08-19 17:36:40 +02:00
Axel Burri
13887837b1
btrbk: add lz4 compression
2016-08-19 16:58:53 +02:00
Axel Burri
3845fa5ec5
btrbk: use format from %compression hash in $raw_postfix_match
2016-08-19 16:58:53 +02:00
Axel Burri
05c54a2390
btrbk: add compression options: pigz, pbzip2, lzo
2016-08-19 16:58:53 +02:00
Axel Burri
a019d8a3e3
btrbk: add configuration option "stream_compress"
...
- add sophisticated stream compression in run_cmd
- add special "compress" cmd_pipe item
- add special "redirect" cmd_pipe item:
use shell redirection instead of troublesome "dd of=".
- disable ssh_compression if stream_compression is set
2016-08-19 16:56:38 +02:00
Axel Burri
99ff88f82a
btrbk: use (and propagate) CONFIG in vinfo instead of hard-setting RSH
2016-08-19 16:54:49 +02:00
Axel Burri
079b947cd6
btrbk: add function WARN_ONCE
2016-08-19 16:48:45 +02:00
Tobias Grosser
84a5c56a83
btrbk: disable compression in gpg
...
We already perform compression before gpg, such that compressing in gpg
is just a waste of time. Interestingly, it seems gpg is not trying to
recompress gzip[ed] input streams, as for the default gzip compression
this patch does not change performance. However, it is necessary for
the upcoming lz4 compression to show its real benefit.
2016-08-07 13:40:45 +02:00
Axel Burri
9a5e8311be
change version to 0.23.3
2016-07-14 01:41:45 +02:00
Axel Burri
796b6bd9bf
btrbk: replace "realpath -e" with "readlink -e" for mountpoint discovery
2016-06-18 13:04:43 +02:00
Axel Burri
91d02ca97b
change version to 0.23.2
2016-06-07 16:41:41 +02:00
Axel Burri
e4a7ba4be5
btrbk: add lockfile support, using perl "flock": add configuration option "lockfile <file>", and command line option "--lockfile"'
2016-06-07 16:17:02 +02:00
Axel Burri
7326b9816c
btrbk: bugfix: raw targets: correctly handle multiple backups in same target directory
2016-05-09 12:42:04 +02:00
Axel Burri
cb0851c8b6
change version to 0.23.1
2016-05-03 16:21:34 +02:00
Axel Burri
5cd03a8fbf
btrbk: remove absolute path to executables, this caused trouble on some distros
2016-05-03 14:34:04 +02:00
Axel Burri
54b5368309
btrbk: code cleanup (remove unneeded code, rename vars)
2016-05-03 13:19:42 +02:00
Axel Burri
12d435d42a
btrbk: explain "orphaned" status in "stats" command, and suppress it on "list backups" command
2016-04-30 13:01:12 +02:00
Axel Burri
46f1e5e2e4
btrbk: only load Data::Dumper if needed (log level "trace", or development version); cosmetics
2016-04-28 13:03:15 +02:00
Axel Burri
49ae46d51d
btrbk: do not print headers to transaction_log, these are documented in btrbk.conf(5)
2016-04-28 12:51:00 +02:00
Axel Burri
e9e398af37
Merge branch 'syslog'
2016-04-28 12:45:13 +02:00
Axel Burri
7e7c28f8f1
btrbk: add eval_quiet(), a simple wrapper around eval, disabling $SIG_(__DIE__)
2016-04-28 12:43:32 +02:00
Axel Burri
082754173c
btrbk: bugfix: raw targets: do not inject child if add_btrbk_filename_info() failed
2016-04-25 21:12:33 +02:00
Axel Burri
47eb14f5fb
btrbk: syslog: add syslog hack in print_formatted(); remove print-to-scalar hack
2016-04-25 19:40:11 +02:00
Axel Burri
c424d917b6
btrbk: syslog: only load Sys::Syslog module if configuration option is set; soft-fail on syslog calls
2016-04-25 18:36:54 +02:00
Hans van Kranenburg
59d96420c8
btrbk: add transaction logging to syslog
...
Add configuration option transaction_syslog, which can be set to a short
name of a syslog facility, like user or local5. Most of the ones besides
localX do not really make sense, but whatever, let the user decide.
The only logging that is relevant for logging to syslog is the logging
generated inside sub action, so it's easy to hijack all messages in
there and also send them to syslog if needed.
All output is done via print_formatted, which expects a file handle.
So, abuse a file handle to a string to be able to change as less code as
needed for this feature.
Since syslog already adds the timestamps for us, I added a syslog
formatting pattern, which is very similar to tlog, omitting the
timestap.
2016-04-25 17:32:08 +02:00
Axel Burri
89df601ee5
btrbk: allow "<hostname>:<path>" notation for volume and target sections
2016-04-25 16:34:52 +02:00
Axel Burri
702ee8950e
btrbk: bugfix: vinfo(): correctly parse url, using check_url()
2016-04-25 15:10:00 +02:00
Axel Burri
357b72bd3f
btrbk: code cleanup: check_file() and check_url() do all the sanitize parts
2016-04-25 14:23:15 +02:00
Axel Burri
191284cd43
btrbk: bugfix: set correct parent section when propagating targets
2016-04-24 15:59:17 +02:00
Axel Burri
4ffaf430c5
change version to 0.23.0
2016-04-23 15:06:57 +02:00
Axel Burri
2e8e4ddefc
btrbk: cosmetics: change info log of send/receive; change topic line; fix typos
2016-04-23 14:58:08 +02:00
Axel Burri
efc260ef91
btrbk: add sanity check on timezone offset; add documentation about implications of long-iso time format
2016-04-23 14:49:12 +02:00
Axel Burri
c7dd680537
btrbk: make sure that snapshots/backups with date in the future are never deleted
2016-04-22 20:51:31 +02:00
Axel Burri
ecb2307b22
btrbk: send/receive: only inject child if target creation was successful
2016-04-22 20:25:30 +02:00
Axel Burri
3ffa63daba
btrbk: bugfix: btrfs_subvolume_show(): quote real_path everywhere
2016-04-22 20:17:26 +02:00
Axel Burri
c13c99ada5
btrbk: add new time_format "long-iso", with seconds and timezone offset (iso8601 format); add function timestamp(): remove dependency to POSIX
2016-04-22 19:47:01 +02:00
Axel Burri
e824c21f50
btrbk: use build-in Time::Local in combination with localtime() instead of Date::Calc; correctly calculate leap hours
2016-04-20 22:45:11 +02:00
Axel Burri
c73dffbd25
btrbk: handle missing Data::Dumper gracefully
2016-04-19 22:12:08 +02:00
Axel Burri
829490f963
btrbk: put btrbk filename info directly to tree node (parse only once)
...
performance win: factor 2.3
2016-04-19 22:04:07 +02:00
Axel Burri
c225231742
btrbk: reuse subvol list for get_receive_targets()
...
performance win: factor 2.6 !
2016-04-19 22:01:23 +02:00
Axel Burri
2b22524ccd
btrbk: sane defaults: xxx_preserve_min defaults to "all", xxx_preserve default to "no" (for backwards compatibility with 0.22.0 default)
2016-04-19 21:37:04 +02:00
Axel Burri
29b9c8fc60
btrbk: action "archive": abort if receive targets of archive candidates exist at unexpected location. makes sure we dont have duplicates on our archives
2016-04-19 21:02:51 +02:00
Axel Burri
5e7a6e5ef4
btrbk: action "archive": delete archives according to archive retention policy
2016-04-19 21:02:51 +02:00
Axel Burri
31f0f97ff0
btrbk: fix info logging for send/receive
2016-04-19 21:02:51 +02:00
Axel Burri
7807fface8
btrbk: cosmetics; s/config_override_opts/config_override_cmdline/g
2016-04-19 21:02:51 +02:00
Axel Burri
2c3997ad0a
btrbk: bugfix: action "config print": correctly handle undef options; skip deprecated options
2016-04-19 21:02:47 +02:00
Axel Burri
a27f99af46
change version to 0.23.0-rc1
2016-04-17 21:46:56 +02:00
Axel Burri
bbb18f5af7
btrbk: cosmetics on info warnings
2016-04-17 21:46:53 +02:00
Axel Burri
edfebb8193
btrbk: action "origin": use --format options; do not recurse into parent chain
2016-04-16 22:05:26 +02:00
Axel Burri
1151d2c572
btrbk: print_formatted(): honor use_header argument for all table formats
2016-04-16 20:25:22 +02:00
Axel Burri
9afe17f738
btrbk: continue gracefully (skip instead of abort) when resuming subvolume having existing (possibly garbled) target; cosmetics on debug/trace log
2016-04-16 19:31:29 +02:00
Axel Burri
ec9dd761b8
btrbk: use "[subvol cmd]" instead of "[btrfs cmd]", and print this as info log
2016-04-16 18:06:55 +02:00
Axel Burri
689d3d1233
btrbk: action "archive": print created subdirectories on summary; cleanup
2016-04-16 17:31:14 +02:00
Axel Burri
5a06a85619
btrbk: moved get_receive_targets_fsroot() out of get_receive_targets(), while cleaning up
2016-04-16 17:25:53 +02:00
Axel Burri
774e6ef842
btrbk: macro_archive_target(): only warn on unexpected targets, not abort
2016-04-16 17:25:53 +02:00
Axel Burri
de05b99757
btrbk: rename action "clone" to "archive" (should have been like this from the beginning)
2016-04-16 17:25:51 +02:00
Axel Burri
4bd68a2e35
btrbk: action "clone": abort all as soon as one target aborts
2016-04-16 00:45:16 +02:00
Axel Burri
cc20dfb8c3
btrbk: schedule(): use "informative_only" option instead of checking for empty value; make sure real values are always in front
2016-04-15 23:29:47 +02:00
Axel Burri
d0cfba7914
btrbk: action "clone": create missing directories
2016-04-15 23:29:47 +02:00
Axel Burri
a5c7c53a86
btrbk: action "clone": print scheduler results if --print-schedule option is set
2016-04-15 23:29:47 +02:00
Axel Burri
14de16aabe
btrbk: add action "clone": recursively send/receive all backups to an archive directory
2016-04-15 23:29:47 +02:00
Axel Burri
3c9aff4c40
btrbk: enable --override option
2016-04-15 11:10:12 +02:00
Axel Burri
e3702305e6
btrbk: bugfix: update uuid_cache after btrfs_subvolume_show()
2016-04-15 11:07:26 +02:00
Axel Burri
6e997674fc
btrbk: remove "btrfs_progs_compat" configuration option, from now on we REQUIRE btrfs-progs >= v3.18.2: we rely on received_uuid for most operations
2016-04-15 11:07:16 +02:00
Axel Burri
6aa54b2dd5
btrbk: bugfix: correctly handle NODE_SUBDIR in vinfo_inject_child()
2016-04-14 18:24:11 +02:00
Axel Burri
141b70f26b
btrbk: bugfix for subdir resolving: on older kernels, the "subvolid" mount option is not shown: try to read it with btrfs_subvolume_show()
2016-04-14 18:21:00 +02:00
Axel Burri
a1ee9d5c6d
btrbk: allow "{snapshot,target}_preserve no": disable retention policy
2016-04-14 14:43:43 +02:00
Axel Burri
a03c515b0a
btrbk: enter failsafe mode on deprecated preserve configuration: at least create the snapshots, and print message to read the upgrade documentation
2016-04-14 13:41:53 +02:00
Axel Burri
7c0987bd3c
btrbk: be harsh (abort) on deprecated "{snapshot,target}_preserve_{daily,weekly,monthly}" and "resume_missing no" options; change default of snapshot_preserve_min to one day
2016-04-13 22:05:20 +02:00
Axel Burri
36cc96fdb8
btrbk: always preserve latest snapshot/backup pair (instead of latest snapshot and latest backup); remove preserve_latest option of schedule()
2016-04-13 22:05:20 +02:00
Axel Burri
032642751b
btrbk: change preserve table text: print latest match instead of earliest match
2016-04-13 22:05:20 +02:00
Axel Burri
e200975f32
btrbk: add option "*preserve_min latest"; deny "snapshot_preserve_min no"
2016-04-13 22:05:20 +02:00
Axel Burri
d143d75b43
btrbk: rename *preserve_all to *preserve_min
2016-04-13 22:05:20 +02:00
Axel Burri
55b121c372
btrbk: remove resume_missing configuration option
2016-04-13 22:05:20 +02:00
Axel Burri
738178565c
btrbk: fix vinfo_inject_child() for raw targets
2016-04-13 22:05:20 +02:00
Axel Burri
a76512955a
btrbk: add vinfo_inject_child(): add a custom node to btr_tree, with fake id, uuid, gen and cgen; use to inject created snapshots and receive targets
2016-04-13 17:17:23 +02:00
Axel Burri
bd34d9f689
btrbk: change preserve logic: keep FIRST present of hour/day/... instead of LAST
2016-04-13 17:17:14 +02:00
Axel Burri
7a334097c1
btrbk: cosmetics (logging and summary text about -p option)
2016-04-12 11:47:41 +02:00
Axel Burri
326edfcc29
btrbk: change preserve semantics (incompatible):
...
- add "{snapshot,target}_preserve_all" configuration option
- change semantics of "{snapshot,target}_preserve NNd" to "preserve
latest daily only"
- change default of "preserve daily,monthly" to 0 (was: all)
- add deprecated warning and enter compatibility mode: preserve
everything if deprecated {snapshot,target}_preserve_* options are
encountered
2016-04-12 11:47:28 +02:00
Axel Burri
ceb4dbf19c
btrbk: add --print-schedule command line option
2016-04-09 14:16:14 +02:00
Axel Burri
78bfc5b885
btrbk: bugfix: use "/bin/false" as RSH if no config is available
2016-04-08 13:38:50 +02:00
Axel Burri
a887e89732
btrbk: warn on receive targets at unexpected location (whole filesystem) when resuming missing backups (do not abort anymore);
2016-04-07 14:35:27 +02:00
Axel Burri
6b71b68fcb
btrbk: bugfix: do not run "rm" command with no arguments (on raw targets)
2016-04-06 20:41:36 +02:00
Axel Burri
5d65748ffe
btrbk: call init_transaction_log() at the beginning of every action, instead at global place
2016-04-06 20:27:27 +02:00
Axel Burri
211964f709
btrbk: parse btrbk_basename and btrbk_date in vinfo_child() instead of vinfo_subvol_list()
2016-04-06 20:24:56 +02:00
Axel Burri
bfb556757d
btrbk: get_latest_common(): use all brothers (snapshots with same parent) as candidates (older first, then newer) instead of only older; as last resort, use subvolumes in snapshot_dir matching btrbk file name scheme as candidates
2016-04-06 15:36:54 +02:00
Axel Burri
46e3aae857
btrbk: get_latest_common(): add strategy to find latest common for readonly and received subvolumes; add helper function cmp_date()
2016-04-06 15:36:51 +02:00
Axel Burri
a26e8851b4
btrbk: btr_tree: return already assembled tree if a subvolume is already known in uuid_cache
2016-04-06 15:36:27 +02:00
Axel Burri
ac0002bce8
btrbk: get_receive_targets() also returns matching source_received_uuid
2016-04-06 15:36:27 +02:00
Axel Burri
4b1983378f
btrbk: add useful flags and parse date and basename of btrbk created files in vinfo_subvol_list(); replace parse_filename() calls by usage of new flags
2016-04-06 13:35:37 +02:00
Axel Burri
0bfb78b842
btrbk: remove trailing slash from NODE_SUBDIR
2016-04-03 21:41:41 +02:00
Axel Burri
899ca92026
btrbk: bugfix: untaint raw backup file list read by "find" command
2016-04-03 21:38:19 +02:00
Axel Burri
023004e9cc
btrbk: bugfix: add '=' character to (regression introduced in a7cd4a99
)
2016-04-03 21:09:30 +02:00
Axel Burri
ea5cd06fbc
btrbk: bugfix: get_receive_targets() for BTRFS_PROGS_COMPAT (regression introduced in 207e8868
)
2016-04-03 21:09:30 +02:00
Axel Burri
302400d640
btrbk: bugfix: disallow snapshot_preserve_* in target section
2016-04-03 15:13:16 +02:00
Axel Burri
a7cd4a994c
btrbk: add some extra safety when running shell commands
2016-04-02 14:13:16 +02:00
Axel Burri
86170e6b21
btrbk: bugfix: untaint result of system_realpath()
2016-04-02 14:10:35 +02:00
Axel Burri
79caba4267
btrbk: rename %symlink -> %realpath_cache (returning real_path instead of real_url)
2016-03-31 17:24:24 +02:00
Axel Burri
7ee5778dee
btrbk: bugfix: print correct target on "same backup target error"
2016-03-31 17:24:24 +02:00
Axel Burri
c4073f6c53
btrbk: print target_path instead of target_subvol for "list {target,config}"
2016-03-31 17:24:24 +02:00
Axel Burri
3ada7c174e
btrbk: allow targets to be directories (use mountpoint framework)
2016-03-31 17:24:24 +02:00
Axel Burri
06043cf800
btrbk: add btrfs mountpoint resolving functionality
2016-03-31 17:24:23 +02:00
Axel Burri
de96f267b8
btrbk: propagate target sections to (sub-)subsections (allow target sections everywhere in the config)
2016-03-31 17:24:18 +02:00
Axel Burri
f9ca7504c8
btrbk: add configuration option "rate_limit" (using pv -L)
...
Original patch provided by @janpascal: "Jan-Pascal van Best"
2016-03-23 11:58:35 +01:00
Axel Burri
e6d46e8a02
btrbk: fix regression: correctly handle previously received subvolumes in get_latest_common()
...
regression from: 8819c67502
2016-03-22 19:30:21 +01:00