enkore
692f2c0933
Merge pull request #2608 from enkore/docs/fix-figure-sizes-pdf
...
docs: fix way too small figures in pdf
2017-06-05 01:02:06 +02:00
Marian Beermann
e80c0f7c5e
docs: fix way too small figures in pdf
2017-06-05 01:01:27 +02:00
Marian Beermann
da04aba5c5
docs: internals: feature flags introduction/rationale
2017-06-05 00:41:30 +02:00
Marian Beermann
c427d238f4
docs: internals: amend feature flags
2017-06-05 00:21:04 +02:00
Marian Beermann
f2fd6fc699
docs: internals: cache feature flags
2017-06-05 00:16:48 +02:00
Marian Beermann
36bdc9d15e
internals: rewrite manifest & feature flags
2017-06-05 00:16:48 +02:00
enkore
71495b90e0
Merge pull request #2606 from enkore/docs/internals-rewrite
...
docs: internals editing
2017-06-04 23:01:29 +02:00
enkore
60811fc2ce
Merge pull request #2593 from enkore/issue/2384.1
...
recreate: if single archive is not processed, exit 2
2017-06-04 23:01:20 +02:00
Marian Beermann
19b425a5c8
docs: internals: more HashIndex details
2017-06-04 23:00:42 +02:00
Marian Beermann
cf77aa53d7
docs: internals: fix ASCII art equations
2017-06-04 22:50:17 +02:00
Marian Beermann
d33929a24d
docs: internals: edited HashIndex
2017-06-04 22:44:29 +02:00
Marian Beermann
daae1cc515
docs/internals: layers image; Blimey! it's one pixel off!
2017-06-04 22:23:54 +02:00
Marian Beermann
a3815034e1
docs: internals: terms/glossary TODO
2017-06-04 22:17:59 +02:00
Marian Beermann
9174f84682
docs: internals: delete non-sequitor from repo/segments
2017-06-04 22:17:56 +02:00
Marian Beermann
89d8f54afb
docs: internals: edited obj graph related sections a bit
2017-06-04 22:17:53 +02:00
Marian Beermann
2b0e7bc924
docs: data structures: demingle cache and repo index
2017-06-04 22:17:48 +02:00
enkore
06980525fa
Merge pull request #2604 from enkore/docs/internals-layers
...
docs/internals: layers image + description
2017-06-04 22:15:25 +02:00
Marian Beermann
f1709df8a9
docs/internals: layers image + description
2017-06-04 22:13:55 +02:00
TW
93e9ca0d23
update CHANGES (master) ( #2594 )
...
update CHANGES (master)
2017-06-04 00:18:09 +02:00
TW
5d689193cc
Merge pull request #2597 from borgbackup/revert-2478-faked-debugging
...
Revert "Start fakeroot faked in debug mode - fixes EISDIR issues"
2017-06-03 23:09:34 +02:00
TW
6119a04122
Merge pull request #2596 from ThomasWaldmann/segs-per-dir
...
DEFAULT_SEGMENTS_PER_DIR = 1000
2017-06-03 22:04:46 +02:00
enkore
b06ceb6547
Revert "Start fakeroot faked in debug mode - fixes EISDIR issues"
2017-06-03 22:02:52 +02:00
Thomas Waldmann
ffcf6b76b6
DEFAULT_SEGMENTS_PER_DIR = 1000
...
prettier increments for the directory names.
2017-06-03 21:54:41 +02:00
TW
a2d08e3186
Merge pull request #2575 from edgimar/master
...
patterns: don't recurse with ! / --exclude for pf:, fixes #2509
2017-06-03 21:47:11 +02:00
enkore
d95551736d
Merge pull request #2591 from enkore/issue/1101.design
...
docs: file integrity
2017-06-03 18:17:14 +02:00
enkore
372eb40089
Merge pull request #2590 from enkore/issue/2589
...
serve: add --restrict-to-repository
2017-06-03 18:17:02 +02:00
Mark Edgington
fd0215c3c2
patterns: don't recurse with !/--exclude for path-prefix (pf:)
...
Fixes issue #2509
2017-06-03 12:13:32 -04:00
Marian Beermann
8ad309ae2a
recreate: if single archive is not processed, exit 2
2017-06-03 15:47:01 +02:00
TW
28e4eb4ea3
add a .coafile for coala ( #2592 )
...
add support for using coala, fixes #1366
ignores / disable are set up so that there are not many faults.
we can improve that iteratively.
2017-06-03 15:13:21 +02:00
Marian Beermann
da99ec2fbd
docs: quotas: refer to --restrict-to-repository
2017-06-03 15:07:08 +02:00
Marian Beermann
8dfe2a8080
remote: show path in PathNotAllowed
...
not 100 % sure whether "if old_server" is required, so let's play it safe.
1.0 -> 1.1 server is no problem.
2017-06-03 15:07:08 +02:00
Marian Beermann
07fbba4ee9
serve: add --restrict-to-repository
2017-06-03 15:07:08 +02:00
Marian Beermann
5af66dbb12
cache sync: add more refcount tests
2017-06-03 15:02:27 +02:00
Steve Groesz
2bcbf8144e
Add bountysource badge ( #2581 )
...
add bountysource badge, fixes #2558
2017-06-03 14:40:08 +02:00
Marian Beermann
b8e40fdce6
editing
2017-06-03 13:05:33 +02:00
Marian Beermann
4faaa7d1fa
RepositoryCache: abort on data corruption
2017-06-03 12:27:35 +02:00
Marian Beermann
b544af2af1
RepositoryCache: checksum decrypted cache
2017-06-03 12:23:23 +02:00
Marian Beermann
45ee62e5ea
docs: file integrity
2017-06-03 00:51:34 +02:00
enkore
c77b758e74
Merge pull request #2584 from enkore/issue/1101.integration.repository
...
repository: checksum index and hints
2017-06-02 22:51:41 +02:00
Marian Beermann
54e023c75a
repository: add complementary index corruption test
2017-06-02 21:44:45 +02:00
Marian Beermann
2e067a7ae8
repository: add refcount corruption test
2017-06-02 21:44:45 +02:00
Marian Beermann
f61ee038d0
repository: checksum index and hints
2017-06-02 21:44:45 +02:00
Marian Beermann
5b3667b617
cache sync: macros in all-caps
2017-06-02 19:31:56 +02:00
Marian Beermann
795cdfc9ab
cache sync: move stat initialization to main unpack
2017-06-02 19:30:53 +02:00
Marian Beermann
cb98cb838d
fuse: fix read(2) caching data in metadata cache
...
The OS page cache is responsible for handling this and is much more
empowered to do a good job at that than Borg.
2017-06-02 17:43:15 +02:00
Marian Beermann
67b97f2223
cache sync: cleanup progress handling, unused parameters
2017-06-02 17:43:15 +02:00
Marian Beermann
7f04e00ba2
testsuite: add TestRepositoryCache
2017-06-02 17:43:15 +02:00
Marian Beermann
835b0e5ee0
cache sync/remote: compressed, decrypted cache
2017-06-02 17:43:15 +02:00
Marian Beermann
c786a5941e
CacheSynchronizer: redo as quasi FSM on top of unpack.h
...
This is a (relatively) simple state machine running in the
data callbacks invoked by the msgpack unpacking stack machine
(the same machine is used in msgpack-c and msgpack-python,
changes are minor and cosmetic, e.g. removal of msgpack_unpack_object,
removal of the C++ template thus porting to C and so on).
Compared to the previous solution this has multiple advantages
- msgpack-c dependency is removed
- this approach is faster and requires fewer and smaller
memory allocations
Testability of the two solutions does not differ in my
professional opinion(tm).
Two other changes were rolled up; _hashindex.c can be compiled
without Python.h again (handy for fuzzing and testing);
a "small" bug in the cache sync was fixed which allocated too
large archive indices, leading to excessive archive.chunks.d
disk usage (that actually gave me an idea).
2017-06-02 17:43:15 +02:00
Marian Beermann
bf895950ac
RepositoryCache: limit cache size
...
Unbounded cache size is inacceptable.
I don't see why a full-fledged repository needs to be used here, either,
since this cache requires none of the consistency or durability guarantees
made by it (and bought with a performance impact).
A notable issue is that posix_fadvise is slow (for some reason) on tmpfs,
which could eat 30-35 % of the total CPU time of a cache sync.
2017-06-02 17:43:15 +02:00
