From d58b6ddf28ef07d87f77a10b472637ec7ea159fc Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Antoine=20Beaupr=C3=A9?= Date: Tue, 16 Dec 2014 10:55:03 -0500 Subject: [PATCH] fix reference errors and remove reference to source code --- docs/internals.rst | 171 +++++++++++++++++++++++++++++++-------------- 1 file changed, 120 insertions(+), 51 deletions(-) diff --git a/docs/internals.rst b/docs/internals.rst index ede9fb0ef..585266a9a 100644 --- a/docs/internals.rst +++ b/docs/internals.rst @@ -56,37 +56,81 @@ on the repository. Repository structure -------------------- -|project_name| is a "filesystem based transactional key value store". +|project_name| is a "filesystem based transactional key value +store". It makes extensive use of msgpack_ to store data and, unless +otherwise noted, data is stored in msgpack_ encoded files. -Objects referenced by a key (256bits id/hash) are stored in line in -files (segments) of size approx 5MB in ``repo/data``. They contain : -header size, crc, size, tag, key, data. Tag is either ``PUT``, -``DELETE``, or ``COMMIT``. Segments are built locally, and then -uploaded. Those files are strictly append-only and modified only once. +Objects referenced by a key (256bits id/hash) are stored inline in +files (`segments`) of size approx 5MB in ``repo/data``. They contain: -A segment file is basically a transaction log where each repository -operation is appended to the file. So if an object is written to the -repository a ``PUT`` tag is written to the file followed by the object -id and data. And if an object is deleted a ``DELETE`` tag is appended +* header size +* crc +* size +* tag +* key +* data + +Segments are built locally, and then uploaded. Those files are +strictly append-only and modified only once. + +Tag is either ``PUT``, ``DELETE``, or ``COMMIT``. A segment file is +basically a transaction log where each repository operation is +appended to the file. So if an object is written to the repository a +``PUT`` tag is written to the file followed by the object id and +data. And if an object is deleted a ``DELETE`` tag is appended followed by the object id. A ``COMMIT`` tag is written when a repository transaction is committed. When a repository is opened any ``PUT`` or ``DELETE`` operations not followed by a ``COMMIT`` tag are discarded since they are part of a partial/uncommitted transaction. The manifest is an object with an id of only zeros (32 bytes), that -references all the archives. It contains : version, list of archives, -timestamp, config. Each archive contains: name, id, time. It is the last -object stored, in the last segment, and is replaced each time. +references all the archives. It contains: + +* version +* list of archives +* timestamp +* config + +Each archive contains: + +* name +* id +* time + +It is the last object stored, in the last segment, and is replaced +each time. The archive metadata does not contain the file items directly. Only references to other objects that contain that data. An archive is an -object that contain metadata : version, name, items list, cmdline, -hostname, username, time. Each item represents a file or directory or -symlink is stored as a ``item`` dictionnary that contains: path, list -of chunks, user, group, uid, gid, mode (item type + permissions), -source (for links), rdev (for devices), mtime, xattrs, acl, -bsdfiles. ``ctime`` (change time) is not stored because there is no -API to set it and it is reset every time an inode's metadata is changed. +object that contain metadata: + +* version +* name +* items list +* cmdline +* hostname +* username +* time + +Each item represents a file or directory or +symlink is stored as a ``item`` dictionnary that contains: + +* path +* list of chunks +* user +* group +* uid +* gid +* mode (item type + permissions) +* source (for links) +* rdev (for devices) +* mtime +* xattrs +* acl +* bsdfiles + +``ctime`` (change time) is not stored because there is no API to set +it and it is reset every time an inode's metadata is changed. All items are serialized using msgpack and the resulting byte stream is fed into the same chunker used for regular file data and turned @@ -97,8 +141,11 @@ beyond the ``MAX_OBJECT_SIZE`` barrier of 20MB. A chunk is an object as well, of course, and its id is the hash of its (unencrypted and uncompressed) content. -Hints are stored in a file (``repo/hints``) and contain: version, list of -segments, compact. +Hints are stored in a file (``repo/hints``) and contain: + +* version +* list of segments +* compact Chunks ------ @@ -113,31 +160,55 @@ stored encrypted in the keyfile. Indexes ------- -The chunk lookup index (chunk hash -> reference count, size, ciphered -size ; in file cache/chunk) and the repository index (chunk hash -> -segment, offset ; in file ``repo/index.%d``) are stored in a sort of hash -table, directly mapped in memory from the file content, with only one -slot per bucket, but that spreads the collisions to the following -buckets. As a consequence the hash is just a start position for a linear -search, and if the element is not in the table the index is linearly -crossed until an empty bucket is found. When the table is full at 90% -its size is doubled, when it's empty at 25% its size is halfed. So -operations on it have a variable complexity between constant and linear -with low factor, and memory overhead varies between 10% and 300%. +There are two main indexes: the chunk lookup index and the repository +index. There is also the file chunk cache. -The file chunk cache (file path hash -> age, inode number, size, -mtime_ns, chunks hashes ; in file cache/files) is stored as a python -associative array storing python objects, which generate a lot of -overhead. This takes around 240 bytes per file without the chunk -list, to be compared to at most 64 bytes of real data (depending on data -alignment), and around 80 bytes per chunk hash (vs 32), with a minimum -of ~250 bytes even if only one chunck hash. The inode number is stored -to make sure we distinguish between different files, as a single path -may not be unique accross different archives in different setups. +The chunk lookup index is stored in ``cache/chunk`` and is indexed on +the ``chunk hash``. It contains: -The ``index.%d`` files are random access but those files can be +* reference count +* size +* ciphered size + +The repository index is stored in ``repo/index.%d`` and is also +indexed on ``chunk hash`` and contains: + +* segment +* offset + +The repository index files are random access but those files can be recreated if damaged or lost using ``check --repair``. +Both indexes are stored as hash tables, directly mapped in memory from +the file content, with only one slot per bucket, but that spreads the +collisions to the following buckets. As a consequence the hash is just +a start position for a linear search, and if the element is not in the +table the index is linearly crossed until an empty bucket is +found. When the table is full at 90% its size is doubled, when it's +empty at 25% its size is halfed. So operations on it have a variable +complexity between constant and linear with low factor, and memory +overhead varies between 10% and 300%. + +The file chunk cache is stored in ``cache/files`` and is indexed on +the ``file path hash`` and contains: + +* age +* inode number +* size +* mtime_ns +* chunks hashes + +The inode number is stored to make sure we distinguish between +different files, as a single path may not be unique accross different +archives in different setups. + +The file chunk cache is stored as a python associative array storing +python objects, which generate a lot of overhead. This takes around +240 bytes per file without the chunk list, to be compared to at most +64 bytes of real data (depending on data alignment), and around 80 +bytes per chunk hash (vs 32), with a minimum of ~250 bytes even if +only one chunck hash. + Indexes memory usage -------------------- @@ -158,9 +229,9 @@ Encryption ---------- AES_ is used with CTR mode of operation (so no need for padding). A 64 -bits initialization vector is used, a SHA256_ based HMAC_ is computed +bits initialization vector is used, a `HMAC-SHA256`_ is computed on the encrypted chunk with a random 64 bits nonce and both are stored -in the chunk. The header of each chunk is : ``TYPE(1)` + +in the chunk. The header of each chunk is : ``TYPE(1)`` + ``HMAC(32)`` + ``NONCE(8)`` + ``CIPHERTEXT``. Encryption and HMAC use two different keys. @@ -185,10 +256,8 @@ Key files When initialized with the ``init -e keyfile`` command, |project_name| needs an associated file in ``$HOME/.attic/keys`` to read and write -the repository. As with most crypto code in |project_name|, the format -of those files is defined in `attic/key.py`_. The format is based on -msgpack_, base64 encoding and PBKDF2_ SHA256 encryption, which is -then encoded again in a msgpack_. +the repository. The format is based on msgpack_, base64 encoding and +PBKDF2_ SHA256 encryption, which is then encoded again in a msgpack_. The internal data structure is as follows: @@ -212,9 +281,9 @@ chunk_seed the seed for the buzhash chunking table (signed 32 bit integer) Those fields are encoded using msgpack_. The utf-8-encoded phassphrase -is encrypted with a PBKDF2_ and SHA256_ using 100000 iterations and a +is encrypted with PBKDF2_ and SHA256_ using 100000 iterations and a random 256 bits salt to give us a derived key. The derived key is 256 -bits long. A HMAC_ SHA256_ checksum of the above fields is generated +bits long. A `HMAC-SHA256`_ checksum of the above fields is generated with the derived key, then the derived key is also used to encrypt the above pack of fields. Then the result is stored in a another msgpack_ formatted as follows: