DOC PREVIEW
Berkeley COMPSCI 186 - Crash Recovery

This preview shows page 1-2 out of 5 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 5 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 5 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 5 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

Crash RecoveryCS 186 Fall 2002, Lecture 25R&G - Chapter 20If you are going to be in the loggingbusiness, one of the things that youhave to do is to learn about heavyequipment.Robert VanNatta,Logging History of Columbia CountyReview: The ACID properties••AAtomicity: All actions in the Xact happen, or nonehappen.••CConsistency: If each Xact is consistent, and the DBstarts consistent, it ends up consistent.••IIsolation: Execution of one Xact is isolated fromthat of other Xacts.••DDurability: If a Xact commits, its effects persist.• Question: which ones does the Recovery Managerhelp with?Atomicity & Durability (andalso used for Consistency-relatedrollbacks)Motivation• Atomicity:– Transactions may abort (“Rollback”).• Durability:– What if DBMS stops running? (Causes?)crash! Desired state after systemrestarts:– T1 & T3 should be durable.– T2, T4 & T5 should beaborted (effects not seen).T1T2T3T4T5AbortCommitCommitAssumptions• Concurrency control is in effect.– Strict 2PL, in particular.• Updates are happening “in place”.– i.e. data is overwritten on (deleted from) theactual page copies (not private copies).• Can you think of a simple scheme (requiring nologging) to guarantee Atomicity & Durability?– What happens during normal execution (what isthe minimum lock granularity)?– What happens when a transaction commits?– What happens when a transaction aborts?Buffer Mgmt Plays a Key Role• Force policy – make sure that every update is on diskbefore commit.– Provides durability without REDO logging.– But, can cause poor performance.• No Steal policy – don’t allow buffer-pool frames withuncommited updates to overwrite committed data ondisk.– Useful for ensuring atomicity without UNDO logging.– But can cause poor performance.Of course, there are some nasty details for gettingForce/NoSteal to work…Preferred Policy: Steal/No-Force• This combination is most complicated but allows forhighest performance.• NO FORCE (complicates enforcing Durability)– What if system crashes before a modified page written bya committed transaction makes it to disk?– Write as little as possible, in a convenient place, at committime, to support REDOing modifications.• STEAL (complicates enforcing Atomicity)– What if the Xact that performed udpates aborts?– What if system crashes before Xact is finished?– Must remember the old value of P (to support UNDOingthe write to page P).Buffer Management summaryForceNo ForceNo Steal StealNo REDONo UNDO UNDONo REDO UNDOREDONo UNDOREDOForceNo ForceNo Steal StealSlowestFastestPerformanceImplicationsLogging/RecoveryImplicationsBasic Idea: Logging• Record REDO and UNDO information, for everyupdate, in a log.– Sequential writes to log (put it on a separate disk).– Minimal info (diff) written to log, so multiple updatesfit in a single log page.• Log: An ordered list of REDO/UNDO actions– Log record contains:<XID, pageID, offset, length, old data, new data>– and additional control info (which we’ll see soon).Write-Ahead Logging (WAL)• The Write-Ahead Logging Protocol: Must force the log record for an update before thecorresponding data page gets to disk. Must force all log records for a Xact before commit.(I.e. transaction is not committed until all of its logrecords including its “commit” record are on thestable log.)• #1 (with UNDO info) helps guarantee Atomicity.• #2 (with REDO info) helps guarantee Durability.• This allows us to implement Steal/No-Force• Exactly how is logging (and recovery!) done?– We’ll look at the ARIES algorithms from IBM.WAL & the Log• Each log record has a unique Log SequenceNumber (LSN).– LSNs always increasing.• Each data page contains a pageLSN.– The LSN of the most recent log recordfor an update to that page.• System keeps track of flushedLSN.– The max LSN flushed so far.• WAL: Before page i is written to DBlog must satisfy:pageLSNi ≤ flushedLSNLSNs pageLSNsRAMflushedLSNpageLSNLog recordsflushed to disk“Log tail” in RAMflushedLSNDBLog RecordsprevLSN is the LSN of theprevious log recordwritten by this Xact (sorecords of an Xact form alinked list backwards intime)Possible log record types:• Update, Commit, Abort• Checkpoint (for logmaintainence)• Compensation LogRecords (CLRs)– for UNDO actions• End (end of commit orabort)LSNprevLSNXIDtypelengthpageIDoffsetbefore-imageafter-imageLogRecord fields:updaterecordsonlyOther Log-Related State• Two in-memory tables:• Transaction Table– One entry per currently active Xact.• entry removed when Xact commits or aborts– Contains XID, status (running/committing/aborting),and lastLSN (most recent LSN written by Xact).• Dirty Page Table:– One entry per dirty page currently in buffer pool.– Contains recLSN -- the LSN of the log record whichfirst caused the page to be dirty.The Big Picture: What’s Stored WhereDBData pageseachwith apageLSNXact TablelastLSNstatusDirty Page TablerecLSNflushedLSNRAMLSNprevLSNXIDtypelengthpageIDoffsetbefore-imageafter-imageLogRecordsLOGMaster recordNormal Execution of an Xact• Series of reads & writes, followed by commit orabort.– We will assume that disk write is atomic.• In practice, additional details to deal with non-atomic writes.• Strict 2PL.• STEAL, NO-FORCE buffer management, with Write-Ahead Logging.Transaction Commit• Write commit record to log.• All log records up to Xact’s commit record areflushed to disk.– Guarantees that flushedLSN ≥ lastLSN.– Note that log flushes are sequential, synchronouswrites to disk.– Many log records per log page.• Commit() returns.• Write end record to log.Simple Transaction Abort• For now, consider an explicit abort of a Xact.– No crash involved.• We want to “play back” the log in reverse order,UNDOing updates.– Get lastLSN of Xact from Xact table.– Write an Abort log record before starting to rollbackoperations– Can follow chain of log records backward via the prevLSNfield.– Write a “CLR” (compensation log record) for each undoneoperation.Abort, cont.• To perform UNDO, must have a lock on data!– No problem!• Before restoring old value of a page, write a CLR:– You continue logging while you UNDO!!– CLR has one extra field: undonextLSN• Points to the next LSN to undo (i.e. the prevLSN of the record we’recurrently undoing).– CLR contains REDO info– CLRs never Undone• Undo needn’t be idempotent (>1 UNDO


View Full Document

Berkeley COMPSCI 186 - Crash Recovery

Documents in this Course
Load more
Download Crash Recovery
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Crash Recovery and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Crash Recovery 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?