Don't prune block locations due to missing blocks a startup (#487)

* Don't prune block locations due to missing blocks a startup
This makes the behaviour consistent with FileCas - you can have an index that is not fully backed by data.
Asking for a location that is not backed by data results in getting an empty result back
Also, don't try to GC blocks that are unknown to the block store at the time of snapshot (to avoid removing data that comes in after GatherReferences in GC)

1 files changed, 3 insertions, 57 deletions

diff --git a/src/zenserver/cache/cachedisklayer.cpp b/src/zenserver/cache/cachedisklayer.cpp
index ebefa2e54..3cd725a27 100644
--- a/src/zenserver/cache/cachedisklayer.cpp
+++ b/src/zenserver/cache/cachedisklayer.cpp
@@ -604,8 +604,7 @@ ZenCacheDiskLayer::CacheBucket::OpenLog(const bool IsNew)
 
 	CreateDirectories(m_BucketDir);
 
-	std::unordered_map<uint32_t, uint64_t> BlockSizes =
-		m_BlockStore.Initialize(m_BlocksBasePath, MaxBlockSize, BlockStoreDiskLocation::MaxBlockIndex + 1);
+	m_BlockStore.Initialize(m_BlocksBasePath, MaxBlockSize, BlockStoreDiskLocation::MaxBlockIndex + 1);
 
 	if (std::filesystem::is_regular_file(IndexPath))
 	{
@@ -636,7 +635,6 @@ ZenCacheDiskLayer::CacheBucket::OpenLog(const bool IsNew)
 
 	std::vector<BlockStoreLocation> KnownLocations;
 	KnownLocations.reserve(m_Index.size());
-	std::vector<DiskIndexEntry> BadEntries;
 	for (const auto& Entry : m_Index)
 	{
 		size_t				 EntryIndex = Entry.second;
@@ -649,48 +647,10 @@ ZenCacheDiskLayer::CacheBucket::OpenLog(const bool IsNew)
 			continue;
 		}
 		const BlockStoreLocation& BlockLocation = Location.GetBlockLocation(m_PayloadAlignment);
-
-		auto BlockIt = BlockSizes.find(BlockLocation.BlockIndex);
-		if (BlockIt == BlockSizes.end())
-		{
-			ZEN_WARN("Unknown block {} for entry {} in '{}'", BlockLocation.BlockIndex, Entry.first.ToHexString(), m_BucketDir);
-		}
-		else
-		{
-			uint64_t BlockSize = BlockIt->second;
-			if (BlockLocation.Offset + BlockLocation.Size > BlockSize)
-			{
-				ZEN_WARN("Range is outside of block {} for entry {} in '{}'",
-						 BlockLocation.BlockIndex,
-						 Entry.first.ToHexString(),
-						 m_BucketDir);
-			}
-			else
-			{
-				KnownLocations.push_back(BlockLocation);
-				continue;
-			}
-		}
-
-		DiskLocation NewLocation = Payload.Location;
-		NewLocation.Flags |= DiskLocation::kTombStone;
-		BadEntries.push_back(DiskIndexEntry{.Key = Entry.first, .Location = NewLocation});
+		KnownLocations.push_back(BlockLocation);
 	}
 
-	if (!BadEntries.empty())
-	{
-		m_SlogFile.Append(BadEntries);
-		m_SlogFile.Flush();
-
-		LogEntryCount += BadEntries.size();
-
-		for (const DiskIndexEntry& BadEntry : BadEntries)
-		{
-			m_Index.erase(BadEntry.Key);
-		}
-	}
-
-	m_BlockStore.Prune(KnownLocations);
+	m_BlockStore.CreateMissingBlocks(KnownLocations);
 
 	if (IsNew || LogEntryCount > 0)
 	{
@@ -1558,11 +1518,6 @@ ZenCacheDiskLayer::CacheBucket::CollectGarbage(GcContext& GcCtx)
 		}
 		DeleteCacheKeys.push_back(ChunkHash);
 	});
-	if (DeleteCacheKeys.empty())
-	{
-		ZEN_DEBUG("garbage collect SKIPPED, for '{}', no expired cache keys found", m_BucketDir);
-		return;
-	}
 
 	auto __ = MakeGuard([&]() {
 		if (!DeletedChunks.empty())
@@ -1638,11 +1593,6 @@ ZenCacheDiskLayer::CacheBucket::CollectGarbage(GcContext& GcCtx)
 				WriteBlockTimeUs += ElapsedUs;
 				WriteBlockLongestTimeUs = std::max(ElapsedUs, WriteBlockLongestTimeUs);
 			});
-			if (m_Index.empty())
-			{
-				ZEN_DEBUG("garbage collect SKIPPED, for '{}', container is empty", m_BucketDir);
-				return;
-			}
 
 			BlockStoreState = m_BlockStore.GetReclaimSnapshotState();
 
@@ -1769,10 +1719,6 @@ ZenCacheDiskLayer::CacheBucket::CollectGarbage(GcContext& GcCtx)
 		}
 	}
 
-	if (TotalChunkHashes.empty())
-	{
-		return;
-	}
 	TotalChunkCount = TotalChunkHashes.size();
 
 	std::vector<BlockStoreLocation> ChunkLocations;