andreacavalli/rocksdb
0
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Update documentation for backups and LogData

Browse Source 
Summary: LogData doesn't consume sequence numbers and doesn't increase the count of the write-batch. Also it was discussed that GetLiveFiles will have to be followed by GetSortedWalFiles to get a lossless backup

Test Plan: visual

Reviewers: dhruba, haobo

Reviewed By: dhruba

CC: leveldb

Differential Revision: https://reviews.facebook.net/D12753
This commit is contained in:
Mayank Agarwal 2013-09-05 14:49:59 -07:00
parent 4b785aab05
commit de98c1d9aa
2 changed files with 11 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
include/rocksdb/db.h
View File

@ -235,14 +235,15 @@ class DB {
// Allow compactions to delete obselete files.
virtual Status EnableFileDeletions() = 0;
// GetLiveFiles followed by GetSortedWalFiles can generate a lossless backup
// THIS METHOD IS DEPRECATED. Use the GetTableMetaData to get more
// detailed information on the live files.
// Retrieve the list of all files in the database. The files are
// relative to the dbname and are not absolute paths. This list
// can be used to generate a backup. The valid size of the manifest
// file is returned in manifest_file_size. The manifest file is
// an ever growing file, but only the portion specified
// by manifest_file_size is valid for this snapshot.
// relative to the dbname and are not absolute paths. The valid size of the
// manifest file is returned in manifest_file_size. The manifest file is an
// ever growing file, but only the portion specified by manifest_file_size is
// valid for this snapshot.
virtual Status GetLiveFiles(std::vector<std::string>&,
uint64_t* manifest_file_size) = 0;

9
include/rocksdb/write_batch.h
View File

@ -43,12 +43,13 @@ class WriteBatch {
// If the database contains a mapping for "key", erase it. Else do nothing.
void Delete(const Slice& key);
// Append a blob of arbitrary to the records in this batch. The blob will be
// stored in the transaction log but not in any other file. In particular, it
// will not be persisted to the SST files. When iterating over this
// Append a blob of arbitrary size to the records in this batch. The blob will
// be stored in the transaction log but not in any other file. In particular,
// it will not be persisted to the SST files. When iterating over this
// WriteBatch, WriteBatch::Handler::LogData will be called with the contents
// of the blob as it is encountered. Blobs, puts, deletes, and merges will be
// encountered in the same order in thich they were inserted.
// encountered in the same order in thich they were inserted. The blob will
// NOT consume sequence number(s) and will NOT increase the count of the batch
//
// Example application: add timestamps to the transaction log for use in
// replication.