haeaeae/duplicati
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
4,233 Commits 140 Branches 416 Tags
28ecfc90527a7bccdf3ff671a7bfb6d4e4ae59ea
Commit Graph
124 Commits
Author SHA1 Message Date
Dan Stahr 27c9e15875 Improved database performance and logging 
- Don't use Vacuum() just before PurgeLogData (which performs vacuum() already)
 - Removed unncessary poorly performning query (after optimizing it I found out it's unused)
 - Correctly log calls for ExecuteScalarInt64
 2017-03-23 15:22:58 +00:00
Jonathan Hardison debfffbe82 Word change. 
Changed from loosing to losing.
Losing - cause/suffering loss.
Loosing - to let loose; free from bonds or restraint. / to let go a hold
 2017-02-13 15:36:39 -06:00
Kenneth Skovhede cafb8e6ff1 Fixed issues with reporting the correct error message when enumerating files 2017-01-23 20:53:18 +01:00
Kenneth Skovhede a029890409 Implemented better default error messsages 2017-01-09 11:35:38 +01:00
Kenneth Skovhede 9a849a2b58 Improved support for cross-os restores. 
This fixes #1302
This fixes #1983
 2017-01-06 23:01:54 +01:00
Kenneth Skovhede 76aae5de7b Changed the logic for getting the last-modified timestamp from the database. 
Now we only look at the timestamp from the previous fileset, instead of for the last time the file was seen.
This should improve lookup times, and thus the running time for backups with many files.

This change means that the following sequence of operations will react differently:

- backup
- delete file
- backup
- undelete file
- backup

Prior to this commit, the undeleted file would be seen with the original timestamp and thus can avoid being scanned for changes.
After this commit, the file will be treated as a new file, even if it has the previous timestamp and will thus be scanned for changes.

An additional switch has been introduced, --check-filetime-only that improves the lookup further by not looking at the metadata or filesize, and thus makes a much lighter call to the database, where fewer tables need to be touched.
 2017-01-05 10:03:32 +01:00
Kenneth Skovhede 9a227ada6b Reworked the cleanup process to allow a dangling filelist, such that this dangling entry can be used as the basis for a synthetic filelist. 
Also added an option to disable synthetic filelists.
This fixes #2038.
This fixes #2101
 2016-12-24 11:42:26 +01:00
Kenneth Skovhede 606b835f7a Improved verbose messages when excluding a file based on the size criteria 2016-12-08 11:09:51 +01:00
Kenneth Skovhede be62ff5556 Added check to avoid adding excluded/missing files to the database. 
This fixes #1855.
 2016-12-08 11:00:12 +01:00
Kenneth Skovhede 32423d78b5 Added EndTime value in a few places. 
This fixes #2127
 2016-12-02 00:08:09 +01:00
Kenneth Skovhede cdba1211a5 Fixed integer overflow when calculating maximum metadata size. 
This fixes #2089
 2016-11-08 10:47:02 +01:00
Kenneth Skovhede 1f7110503a Updated unix symlinks handling to store correct metadata for symlinks 2016-11-03 13:11:53 +01:00
Max 5c7dfcada7 Refactor of hyper-v backup 2016-10-22 23:08:27 +02:00
Kenneth SkovhedeandGitHub 431b917ae5 Merge pull request #2020 from AndersGron/master 
Amazon cloud drive consistency delay
 2016-10-18 18:55:30 +02:00
Kenneth Skovhede 3880df7e6d Merge branch 'hyperv_ui' of https://github.com/mnaiman/duplicati into mnaiman-hyperv_ui 
# Conflicts:
#	Duplicati/Library/Main/Operation/BackupHandler.cs
 2016-10-18 13:31:22 +02:00
Kenneth Skovhede 60c972cf0b Added a comment that explains how to get more than 300mb metadata support, should we need it 2016-10-17 15:43:41 +02:00
Kenneth Skovhede 309bf1648d Merge branch 'master' into feature/fix_for_different_blockhash_and_filehash_plus_multiblocks_metadata 
# Conflicts:
#	Duplicati/Library/Main/Database/LocalDatabase.cs
#	Duplicati/Library/Main/Database/LocalRecreateDatabase.cs
#	Duplicati/Library/Main/Operation/RecreateDatabaseHandler.cs
#	Duplicati/UnitTest/BorderTests.cs
 2016-10-17 13:28:27 +02:00
AndersGron b008a54302 Fix use of null object 
Fix use of null object
 2016-10-16 22:02:55 +02:00
Max 06cccea7a7 moved hyper-v code to hyper-v classes 2016-10-16 01:40:12 +02:00
Max 8c72cefd5b changes in hyper-v backed 2016-10-15 23:55:50 +02:00
Max d03c794d0f fixes 3 2016-10-07 15:03:53 +02:00
Max 0cbd292924 fixes for hyper-v 
basic logging
added safe checks
 2016-10-06 00:26:54 +02:00
Kenneth Skovhede a0368d5f30 Added support for reporting backend size when --no-backend-verification is supplied. 
This fixes #1723
 2016-09-23 13:46:42 +02:00
Kenneth Skovhede 0c68daff6c Fixed all whitespace to be 4 spaces instead of tabs 2016-09-15 11:39:27 +02:00
Kenneth Skovhede 490776ed73 Implemented a database flag to mark a database as partially recreated. 
Fixed issue with error messages when attempting to recreate a partial database for restore as described in #1695.
 2016-09-13 21:55:15 +02:00
Kenneth Skovhede 72e3782b0a Added error reporting to detect errors such as the one reported in #1905. 2016-08-30 12:26:57 +02:00
Kenneth Skovhede 3f829cb6c1 Removed errors related to files reporting to be "reparsepoint" but not being symlinks. 
This should fix #1727
 2016-04-27 14:49:56 +02:00
Kenneth Skovhede d101f91c94 Fixed support for metadata larger than blocksize. Now supports one full blocklist of metadata, appx 300mb with default settings. 
Added correct handling of metadata hashes, such that a metadata blockset is now handled the same was as a regular file blockset.
The query that creates the filesets does not support more than a single blocklist of metadata, hence the limitiation.
 2016-04-04 18:11:48 +02:00
Kenneth Skovhede 72b33bcf93 Extra fixes to support larger metadata (up to appx 300 mb with default blocksize) 2016-04-03 23:54:52 +02:00
Kenneth Skovhede e40bd7f02a Proposed version, not complete with handling large metadata chunks 2016-04-03 23:10:47 +02:00
Kenneth Skovhede 875cd57eab Bit too quick there :) 2016-03-20 12:41:17 +01:00
Kenneth Skovhede bf07813ca8 Real fix for the problem with duplicate fileset timestamp on really short backups 2016-03-20 12:38:36 +01:00
Kenneth Skovhede 0530cb4d52 Added repair-in-progress check to avoid continuing work on a broken database 2016-03-18 13:26:12 +01:00
Kenneth Skovhede be959a8a1b Changed the name of a resource class 2016-03-14 09:35:01 +01:00
Kenneth Skovhede 1fab22c807 Fixed some crashes introduced with the refactoring 2016-02-19 00:15:16 +01:00
Kenneth Skovhede 3236459f86 More refactoring to reduce the number of shared variables used when calling around in the BackupHandler class 2016-02-18 23:10:26 +01:00
Kenneth Skovhede 93cf6f6690 Changed logic for creating new volumes during backup. 
Instead of creating them up front, they are created when needed (lazy init).
 2016-02-18 22:49:10 +01:00
Kenneth Skovhede 781d8e4db4 Refactored the main backup routine to be more readable. 
It still sucks because it keeps too much state in the class variables between function calls
 2016-02-18 22:38:03 +01:00
Kenneth Skovhede 4780c25ff0 Bugfix, uploading control files in synthetic filelists did not work 2016-02-18 22:37:45 +01:00
Kenneth Skovhede 9bfc39120a Improved handling the compaction process with errors, such that deletes are marked and commited to the database before actually being deleted, causing fewer issues when resuming the operation afterwards. 2015-11-17 21:32:55 +01:00
Kenneth Skovhede 8359369952 Added error handling to treat errors in reading metadata as warnings instead of skipping the entire file 2015-09-04 13:14:30 +02:00
Kenneth Skovhede e13978d1c0 Improved handling of non-standard blocksize and hashing algorithms 2015-04-08 20:33:30 +02:00
Kenneth Skovhede 0de9adc606 Added grace-period for incomplete uploads to avoid backups stopping due to the Apache WebDAV issues. 
This fixes #1243.
 2015-04-05 14:33:13 +02:00
Kenneth Skovhede 3972211757 Added automatic log cleanup 
This fixes #1298
 2015-02-15 23:44:20 +01:00
Kenneth Skovhede a45b6ea621 Added control files to synthetic dlist files too 2015-02-03 23:49:12 +01:00
Kenneth Skovhede 60389b1e6e Added extra consistency check to detect the missing file issue related to issue #1214. 2015-02-03 23:47:33 +01:00
Kenneth Skovhede 4eefccc4f2 Fixed an issue where changed metadata was not stored if the file data had not changed. 
This also includes a fix for a path-lookup that was larger than needed.
 2015-01-25 21:55:15 +01:00
Kenneth Skovhede ed6acedd67 Fixed a bug where the old modification timestamp was re-recorded, resulting in repeated scanning of the files 2015-01-25 21:51:34 +01:00
Kenneth Skovhede 4846ffa103 Now using the file modification timestamp and the filesize as indicators that trigger file scanning. 
This fixes #1099.
 2015-01-24 21:59:53 +01:00
Kenneth Skovhede 16c791f0b5 Removed all resx files, now using Duplicati.Library.Localization exclusively 2015-01-20 21:07:24 +01:00