Files
duplicati/Duplicati/Library/Main/Operation/CompactHandler.cs
T
Kenneth Skovhede 7f008341df Untangle the results and database
The code was using a two-way reference linking the database to the results and the results to the database.

However, this was not really used so it ended up just cluttering the code. With this commit, the results and databases are no longer keeping track of each other, and the only interaction is when the results are written to the database at the end of the operation.

To ensure result data is always written, this code has been moved into the controller, which will now flush the results for all operations. Prior to this, only some operations would write result data to the log.

The controller now also handles logic for cleaning old log data and invoking the vacuum command.

Fixed a bug due to using System.Text.Json where some attributes from NewtonSoft.Json were being applied and not picked up.
2025-03-20 09:51:49 +01:00

361 lines
21 KiB
C#

// Copyright (C) 2025, The Duplicati Team
// https://duplicati.com, hello@duplicati.com
//
// Permission is hereby granted, free of charge, to any person obtaining a
// copy of this software and associated documentation files (the "Software"),
// to deal in the Software without restriction, including without limitation
// the rights to use, copy, modify, merge, publish, distribute, sublicense,
// and/or sell copies of the Software, and to permit persons to whom the
// Software is furnished to do so, subject to the following conditions:
//
// The above copyright notice and this permission notice shall be included in
// all copies or substantial portions of the Software.
//
// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
// OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
// FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
// DEALINGS IN THE SOFTWARE.
using Duplicati.Library.Main.Database;
using Duplicati.Library.Main.Volumes;
using Duplicati.Library.Utility;
using System;
using System.Collections.Generic;
using System.Data;
using System.Linq;
using System.Runtime.CompilerServices;
using System.Threading;
using System.Threading.Tasks;
namespace Duplicati.Library.Main.Operation
{
internal class CompactHandler
{
/// <summary>
/// The tag used for logging
/// </summary>
private static readonly string LOGTAG = Logging.Log.LogTagFromType<CompactHandler>();
protected readonly Options m_options;
protected readonly CompactResults m_result;
public CompactHandler(Options options, CompactResults result)
{
m_options = options;
m_result = result;
}
public async Task RunAsync(IBackendManager backendManager)
{
if (!System.IO.File.Exists(m_options.Dbpath))
throw new Exception(string.Format("Database file does not exist: {0}", m_options.Dbpath));
using (var db = new LocalDeleteDatabase(m_options.Dbpath, "Compact"))
using (var tr = new ReusableTransaction(db))
{
Utility.UpdateOptionsFromDb(db, m_options);
Utility.VerifyOptionsAndUpdateDatabase(db, m_options);
var changed = await DoCompactAsync(db, false, tr, backendManager).ConfigureAwait(false);
if (changed && m_options.UploadVerificationFile)
await FilelistProcessor.UploadVerificationFile(backendManager, m_options, db, null);
if (!m_options.Dryrun)
{
tr.Commit("CommitCompact", restart: false);
if (changed)
{
db.WriteResults(m_result);
if (m_options.AutoVacuum)
{
m_result.VacuumResults = new VacuumResults(m_result);
await new VacuumHandler(m_options, (VacuumResults)m_result.VacuumResults).RunAsync().ConfigureAwait(false);
}
}
}
}
}
internal async Task<bool> DoCompactAsync(LocalDeleteDatabase db, bool hasVerifiedBackend, ReusableTransaction rtr, IBackendManager backendManager)
{
var report = db.GetCompactReport(m_options.VolumeSize, m_options.Threshold, m_options.SmallFileSize, m_options.SmallFileMaxCount, rtr.Transaction);
report.ReportCompactData();
if (report.ShouldReclaim || report.ShouldCompact)
{
// Workaround where we allow a running backendmanager to be used
if (!hasVerifiedBackend)
await FilelistProcessor.VerifyRemoteList(backendManager, m_options, db, m_result.BackendWriter, true, FilelistProcessor.VerifyMode.VerifyStrict, rtr.Transaction).ConfigureAwait(false);
var newvol = new BlockVolumeWriter(m_options);
newvol.VolumeID = db.RegisterRemoteVolume(newvol.RemoteFilename, RemoteVolumeType.Blocks, RemoteVolumeState.Temporary, rtr.Transaction);
IndexVolumeWriter newvolindex = null;
if (m_options.IndexfilePolicy != Options.IndexFileStrategy.None)
{
newvolindex = new IndexVolumeWriter(m_options);
newvolindex.VolumeID = db.RegisterRemoteVolume(newvolindex.RemoteFilename, RemoteVolumeType.Index, RemoteVolumeState.Temporary, rtr.Transaction);
db.AddIndexBlockLink(newvolindex.VolumeID, newvol.VolumeID, rtr.Transaction);
}
var blocksInVolume = 0L;
var buffer = new byte[m_options.Blocksize];
var remoteList = db.GetRemoteVolumes().Where(n => n.State == RemoteVolumeState.Uploaded || n.State == RemoteVolumeState.Verified).ToArray();
//These are for bookkeeping
var uploadedVolumes = new List<KeyValuePair<string, long>>();
var deletedVolumes = new List<KeyValuePair<string, long>>();
var downloadedVolumes = new List<KeyValuePair<string, long>>();
//We start by deleting unused volumes to save space before uploading new stuff
List<IRemoteVolume> fullyDeleteable = [];
if (report.DeleteableVolumes.Any())
{
var deleteableVolumesAsHashSet = new HashSet<string>(report.DeleteableVolumes);
fullyDeleteable =
remoteList
.Where(n => deleteableVolumesAsHashSet.Contains(n.Name))
.Cast<IRemoteVolume>()
.ToList();
}
await foreach (var d in DoDelete(db, backendManager, fullyDeleteable, rtr, m_result.TaskControl.ProgressToken).ConfigureAwait(false))
deletedVolumes.Add(d);
// This list is used to pick up unused volumes,
// so they can be deleted once the upload of the
// required fragments is complete
var deleteableVolumes = new List<IRemoteVolume>();
if (report.ShouldCompact)
{
// If we crash now, we may leave partial files
if (!m_options.Dryrun)
db.TerminatedWithActiveUploads = true;
newvolindex?.StartVolume(newvol.RemoteFilename);
List<IRemoteVolume> volumesToDownload = [];
if (report.CompactableVolumes.Any())
{
var compactableVolumesAsHashSet = new HashSet<string>(report.CompactableVolumes);
volumesToDownload =
remoteList
.Where(n => compactableVolumesAsHashSet.Contains(n.Name))
.Cast<IRemoteVolume>()
.ToList();
}
using (var q = db.CreateBlockQueryHelper(rtr.Transaction))
{
await foreach (var (tmpfile, hash, size, name) in backendManager.GetFilesOverlappedAsync(volumesToDownload, m_result.TaskControl.ProgressToken).ConfigureAwait(false))
{
using (tmpfile)
{
var entry = new RemoteVolume(name, hash, size);
if (!await m_result.TaskControl.ProgressRendevouz().ConfigureAwait(false))
{
await backendManager.WaitForEmptyAsync(db, rtr.Transaction, m_result.TaskControl.ProgressToken).ConfigureAwait(false);
return false;
}
downloadedVolumes.Add(new KeyValuePair<string, long>(entry.Name, entry.Size));
var inst = VolumeBase.ParseFilename(entry.Name);
using (var f = new BlockVolumeReader(inst.CompressionModule, tmpfile, m_options))
{
foreach (var e in f.Blocks)
{
if (q.UseBlock(e.Key, e.Value, rtr.Transaction))
{
//TODO: How do we get the compression hint? Reverse query for filename in db?
var s = f.ReadBlock(e.Key, buffer);
if (s != e.Value)
throw new Exception(string.Format("Size mismatch problem for block {0}, {1} vs {2}", e.Key, s, e.Value));
newvol.AddBlock(e.Key, buffer, 0, s, Interface.CompressionHint.Compressible);
if (newvolindex != null)
newvolindex.AddBlock(e.Key, e.Value);
db.RegisterDuplicatedBlock(e.Key, e.Value, newvol.VolumeID, rtr.Transaction);
blocksInVolume++;
if (newvol.Filesize > (m_options.VolumeSize - m_options.Blocksize))
{
await FinishVolumeAndUpload(db, backendManager, newvol, newvolindex, uploadedVolumes);
newvol = new BlockVolumeWriter(m_options);
newvol.VolumeID = db.RegisterRemoteVolume(newvol.RemoteFilename, RemoteVolumeType.Blocks, RemoteVolumeState.Temporary, rtr.Transaction);
if (m_options.IndexfilePolicy != Options.IndexFileStrategy.None)
{
newvolindex = new IndexVolumeWriter(m_options);
newvolindex.VolumeID = db.RegisterRemoteVolume(newvolindex.RemoteFilename, RemoteVolumeType.Index, RemoteVolumeState.Temporary, rtr.Transaction);
db.AddIndexBlockLink(newvolindex.VolumeID, newvol.VolumeID, rtr.Transaction);
newvolindex.StartVolume(newvol.RemoteFilename);
}
blocksInVolume = 0;
// Wait for the backend to catch up
await backendManager.WaitForEmptyAsync(db, rtr.Transaction, m_result.TaskControl.ProgressToken).ConfigureAwait(false);
// Commit as we have uploaded a volume
if (!m_options.Dryrun)
rtr.Commit();
}
}
}
}
deleteableVolumes.Add(entry);
}
}
if (blocksInVolume > 0)
{
await FinishVolumeAndUpload(db, backendManager, newvol, newvolindex, uploadedVolumes).ConfigureAwait(false);
}
else
{
db.RemoveRemoteVolume(newvol.RemoteFilename, rtr.Transaction);
if (newvolindex != null)
{
db.RemoveRemoteVolume(newvolindex.RemoteFilename, rtr.Transaction);
newvolindex.FinishVolume(null, 0);
}
}
}
// The remainder of the operation cannot leave partial files
if (!m_options.Dryrun)
db.TerminatedWithActiveUploads = false;
}
else
{
newvolindex?.Dispose();
newvol.Dispose();
}
await foreach (var d in DoDelete(db, backendManager, deleteableVolumes, rtr, m_result.TaskControl.ProgressToken).ConfigureAwait(false))
deletedVolumes.Add(d);
var downloadSize = downloadedVolumes.Where(x => x.Value >= 0).Aggregate(0L, (a, x) => a + x.Value);
var deletedSize = deletedVolumes.Where(x => x.Value >= 0).Aggregate(0L, (a, x) => a + x.Value);
var uploadSize = uploadedVolumes.Where(x => x.Value >= 0).Aggregate(0L, (a, x) => a + x.Value);
m_result.DeletedFileCount = deletedVolumes.Count;
m_result.DownloadedFileCount = downloadedVolumes.Count;
m_result.UploadedFileCount = uploadedVolumes.Count;
m_result.DeletedFileSize = deletedSize;
m_result.DownloadedFileSize = downloadSize;
m_result.UploadedFileSize = uploadSize;
m_result.Dryrun = m_options.Dryrun;
if (m_result.Dryrun)
{
if (downloadedVolumes.Count == 0)
Logging.Log.WriteDryrunMessage(LOGTAG, "CompactResults", "Would delete {0} files, which would reduce storage by {1}", m_result.DeletedFileCount, Library.Utility.Utility.FormatSizeString(m_result.DeletedFileSize));
else
Logging.Log.WriteDryrunMessage(LOGTAG, "CompactResults", "Would download {0} file(s) with a total size of {1}, delete {2} file(s) with a total size of {3}, and compact to {4} file(s) with a size of {5}, which would reduce storage by {6} file(s) and {7}",
m_result.DownloadedFileCount,
Library.Utility.Utility.FormatSizeString(m_result.DownloadedFileSize),
m_result.DeletedFileCount,
Library.Utility.Utility.FormatSizeString(m_result.DeletedFileSize), m_result.UploadedFileCount,
Library.Utility.Utility.FormatSizeString(m_result.UploadedFileSize),
m_result.DeletedFileCount - m_result.UploadedFileCount,
Library.Utility.Utility.FormatSizeString(m_result.DeletedFileSize - m_result.UploadedFileSize));
}
else
{
if (m_result.DownloadedFileCount == 0)
Logging.Log.WriteInformationMessage(LOGTAG, "CompactResults", "Deleted {0} files, which reduced storage by {1}", m_result.DeletedFileCount, Library.Utility.Utility.FormatSizeString(m_result.DeletedFileSize));
else
Logging.Log.WriteInformationMessage(LOGTAG, "CompactResults", "Downloaded {0} file(s) with a total size of {1}, deleted {2} file(s) with a total size of {3}, and compacted to {4} file(s) with a size of {5}, which reduced storage by {6} file(s) and {7}",
m_result.DownloadedFileCount,
Library.Utility.Utility.FormatSizeString(downloadSize),
m_result.DeletedFileCount,
Library.Utility.Utility.FormatSizeString(m_result.DeletedFileSize),
m_result.UploadedFileCount,
Library.Utility.Utility.FormatSizeString(m_result.UploadedFileSize),
m_result.DeletedFileCount - m_result.UploadedFileCount,
Library.Utility.Utility.FormatSizeString(m_result.DeletedFileSize - m_result.UploadedFileSize));
}
await backendManager.WaitForEmptyAsync(db, rtr.Transaction, m_result.TaskControl.ProgressToken).ConfigureAwait(false);
m_result.EndTime = DateTime.UtcNow;
return (m_result.DeletedFileCount + m_result.UploadedFileCount) > 0;
}
else
{
m_result.EndTime = DateTime.UtcNow;
return false;
}
}
private async IAsyncEnumerable<KeyValuePair<string, long>> DoDelete(LocalDeleteDatabase db, IBackendManager backend, IEnumerable<IRemoteVolume> deleteableVolumes, ReusableTransaction rtr, [EnumeratorCancellation] CancellationToken cancellationToken)
{
// Find volumes that can be deleted
var remoteFilesToRemove = db.ReOrderDeleteableVolumes(deleteableVolumes, rtr.Transaction).ToList();
// Make sure we do not re-assign blocks to any of the volumes we are about to delete
var toRemoveVolumeIds = db.GetRemoteVolumeIDs(remoteFilesToRemove.Select(x => x.Name), rtr.Transaction)
.Select(x => x.Value)
.Distinct()
.ToList();
// Mark all volumes and relevant index files as disposable
foreach (var f in remoteFilesToRemove)
{
db.PrepareForDelete(f.Name, toRemoveVolumeIds, rtr.Transaction);
db.UpdateRemoteVolume(f.Name, RemoteVolumeState.Deleting, f.Size, f.Hash, rtr.Transaction);
}
// Before we commit the current state, make sure the backend has caught up
await backend.WaitForEmptyAsync(db, rtr.Transaction, cancellationToken).ConfigureAwait(false);
if (!m_options.Dryrun)
rtr.Commit();
await foreach (var d in PerformDelete(backend, remoteFilesToRemove, cancellationToken).ConfigureAwait(false))
yield return d;
}
private async Task FinishVolumeAndUpload(LocalDeleteDatabase db, IBackendManager backendManager, BlockVolumeWriter newvol, IndexVolumeWriter newvolindex, List<KeyValuePair<string, long>> uploadedVolumes)
{
Action indexVolumeFinished = () =>
{
if (newvolindex != null && m_options.IndexfilePolicy == Options.IndexFileStrategy.Full)
{
foreach (var blocklist in db.GetBlocklists(newvol.VolumeID, m_options.Blocksize, m_options.BlockhashSize))
{
newvolindex.WriteBlocklist(blocklist.Item1, blocklist.Item2, 0, blocklist.Item3);
}
}
};
uploadedVolumes.Add(new KeyValuePair<string, long>(newvol.RemoteFilename, newvol.Filesize));
if (newvolindex != null)
uploadedVolumes.Add(new KeyValuePair<string, long>(newvolindex.RemoteFilename, newvolindex.Filesize));
if (!m_options.Dryrun)
await backendManager.PutAsync(newvol, newvolindex, indexVolumeFinished, false, m_result.TaskControl.ProgressToken).ConfigureAwait(false);
else
Logging.Log.WriteDryrunMessage(LOGTAG, "WouldUploadGeneratedBlockset", "Would upload generated blockset of size {0}", Library.Utility.Utility.FormatSizeString(newvol.Filesize));
}
private async IAsyncEnumerable<KeyValuePair<string, long>> PerformDelete(IBackendManager backendManager, IEnumerable<IRemoteVolume> list, [EnumeratorCancellation] CancellationToken cancelToken)
{
foreach (var f in list)
{
if (!m_options.Dryrun)
await backendManager.DeleteAsync(f.Name, f.Size, false, cancelToken).ConfigureAwait(false);
else
Logging.Log.WriteDryrunMessage(LOGTAG, "WouldDeleteRemoteFile", "Would delete remote file: {0}, size: {1}", f.Name, Library.Utility.Utility.FormatSizeString(f.Size));
yield return new KeyValuePair<string, long>(f.Name, f.Size);
}
}
}
}