improved scrubbing of oplogs and filecas (#596)

- Improvement: Scrub command now validates compressed buffer hashes in filecas storage (used for large chunks)
- Improvement: Added --dry, --no-gc and --no-cas options to zen scrub command
- Improvement: Implemented oplog scrubbing (previously was a no-op)
- Improvement: Implemented support for running scrubbint at startup with --scrub=<options>

2 files changed, 8 insertions, 1 deletions

diff --git a/src/zenstore/include/zenstore/gc.h b/src/zenstore/include/zenstore/gc.h
index 698b0d4e8..30dd97ce8 100644
--- a/src/zenstore/include/zenstore/gc.h
+++ b/src/zenstore/include/zenstore/gc.h
@@ -492,6 +492,8 @@ public:
 	{
 		bool				 SkipGc		  = false;
 		std::chrono::seconds MaxTimeslice = std::chrono::seconds::max();
+		bool				 SkipDelete	  = false;
+		bool				 SkipCas	  = false;
 	};
 
 	bool TriggerScrub(const TriggerScrubParams& Params);
@@ -508,7 +510,7 @@ private:
 								GcVersion				  UseGCVersion,
 								uint32_t				  CompactBlockUsageThresholdPercent,
 								bool					  Verbose);
-	void		 ScrubStorage(bool DoDelete, std::chrono::seconds TimeSlice);
+	void		 ScrubStorage(bool DoDelete, bool SkipCid, std::chrono::seconds TimeSlice);
 	LoggerRef	 Log() { return m_Log; }
 	virtual bool AreDiskWritesAllowed() const override { return !m_AreDiskWritesBlocked.load(); }
 	DiskSpace	 CheckDiskSpace();
diff --git a/src/zenstore/include/zenstore/scrubcontext.h b/src/zenstore/include/zenstore/scrubcontext.h
index cefaf0888..2f28cfec7 100644
--- a/src/zenstore/include/zenstore/scrubcontext.h
+++ b/src/zenstore/include/zenstore/scrubcontext.h
@@ -38,15 +38,20 @@ public:
 	inline uint64_t ScrubbedBytes() const { return m_ByteCount; }
 
 	HashKeySet BadCids() const;
+	bool	   IsBadCid(const IoHash& Cid) const;
 
 	inline bool RunRecovery() const { return m_Recover; }
 	inline void SetShouldDelete(bool DoDelete) { m_Recover = DoDelete; }
 
+	inline bool IsSkipCas() const { return m_SkipCas; }
+	inline void SetSkipCas(bool DoSkipCas) { m_SkipCas = DoSkipCas; }
+
 	inline WorkerThreadPool& ThreadPool() { return m_WorkerThreadPool; }
 
 private:
 	uint64_t							  m_ScrubTime = GetHifreqTimerValue();
 	bool								  m_Recover	  = true;
+	bool								  m_SkipCas	  = false;
 	std::atomic<uint64_t>				  m_ChunkCount{0};
 	std::atomic<uint64_t>				  m_ByteCount{0};
 	mutable RwLock						  m_Lock;