Resource Managers Jim Gray Microsoft Gray Microsoft com

Resource Managers Jim Gray Microsoft, Gray @ Microsoft. com Andreas Reuter International University, Andreas. Reuter@i-u. de Mon Tue Wed Thur Fri 9: 00 Overview TP mons Log Files &Buffers B-tree 11: 00 Faults Lock Theory Res. Mgr COM+ Access Paths 1: 30 Tolerance Lock Techniq CICS & Inet Corba Groupware 3: 30 T Models Queues Adv TM Replication Benchmark 7: 00 Party Workflow Cyberbrick Party Gray & Reuter: Resource Manager 1

Whirlwind Tour: The Actors Resource managers – – provide ACID objects (transactional objects) Use log manager to record changes Use transaction manager to coordinate multi-RM changes Use communication manager to make transactional RPCs Communication Manager Resource Managers Objects Communication Manager Transaction Manager Log Gray & Reuter: Resource Manager Objects Volatile Storage Durable Storage Resource Managers Durable Storage 2

Whirlwind Tour: the Application Verbs TRID Begin_Work(context *); /* begin a transaction */ Boolean Commit_Work(context *); /* commit the transaction void Abort_Work(void); /* rollback to savepoint zero */ */ savepoint Save_Work(context *); /* establish a savepoint */ savepoint Rollback_Work(savepoint); /*return to savept (savept 0 = abort)*/ Boolean Prepare_Work(context *); /* put transaction in prepared state */ context Read_Context(void); /* return current savepoint context */ TRID Chain_Work(context *); /* end current and start next trans */ TRID My_Trid(void); /* return current transaction identifier*/ TRID Leave_Transaction(void); /*set process trid null, return current id*/ Boolean Resume_Transaction(TRID); /* set process trid to desired trid */ enum tran_status { ACTIVE , PREPARED , ABORTING , COMMITTING , ABORTED , COMMITTED}; tran_status Status_Transaction(TRID); /* transaction identifier status */ Gray & Reuter: Resource Manager 3

Whirlwind Tour Types Of Transaction Executions A Simple Commit Begin Action Save Action Commit A Simple Abort Begin Action Save Action Rollback A Partial Rollback Begin Action Save Action Rollback A Persistent Transaction Surviving A System Restart Action Save Action Commit Begin Action Save Persistent Action Save Action Restart Action Save Action Commit Shaded stuff is “undone” Gray & Reuter: Resource Manager 4

Whirlwind Tour: the TRID Flow Call graph: who calls whom. TRIDs flow on all such calls. Application is typically root. RM can be an application (use a transactional RM to store state) Transaction Application Servers Resource Managers Application Gray & Reuter: Resource Manager 5

Whirlwind tour Normal (no failure) Transaction Execution TM generates the TRID at Begin_Work(). Coordinates Commit, RM joins work, generates log records, allows commit Gray & Reuter: Resource Manager 6

WW tour: The Resource Manger view Transaction Manager Identify Save. Work Rollback. Work Join Status. Transaction Leave Resume Save Prepare Commit UNDO REDO Checkpoint rm. Call(. . . ) response TP monitor administrative functions and callbacks to install, start, and schedule a resource manager's own service interface functions transaction management callbacks Gray & Reuter: Resource Manager invocation other resource managers callbacks rm. Call(. . . ) (depends on application) 7

WW tour: The Resource manager view Boolean */ Boolean void Savepoint(LSN *); /* invoked at tran Save_Work(). Returns RM vote void Boolean void UNDO(LSN); /* Undo the log record with this LSN */ REDO(LSN); /* Redo the log record with this LSN */ UNDO_Savepoint(LSN); /* Vote TRUE if can return to savepoint REDO_Savepoint(LSN); /* Redo a savepoint. */ Prepare(LSN *); /* invoked at phase_1. Return vote on commit Commit(); /* called at commit ¯ 2 */ Abort(); /* called at failed commit ¯ 2 or abort */ */ */ void LSN TM_Startup(LSN); /* TM restarting. Passes RM ckpt LSN */ Checkpoint(LSN * low_water); /* TM checkpointing, Return RM ckpt LSN, set low water LSN */ Boolean Join_Work(RMID, TRID); /* Become part of a transaction */ Gray & Reuter: Resource Manager 8

WW Tour: The Transaction Manager Transaction rollback. coordinates transaction rollback to a savepoint or abort rollbacks can be initiated by any participant. Resource manager restart. If an RM fails and restarts, TM presents checkpoint anchor & RM undo/redo log System restart. TM drives local RM recovery (like RM restart) TM resolves any in-doubt distributed transactions Media recovery. TM helps RM reconstruct damaged objects by providing archive copies of object + the log of object since archived. Node restart. Transaction commit among independent TMs when a TM fails. Gray & Reuter: Resource Manager 9

WW Tour: When a Transaction Aborts At transaction rollback TM drives undo of each RM joined to the transaction Can be to savepoint 0 (abort) or partial rollback. Gray & Reuter: Resource Manager 10

WW tour: the Transaction Manager at Restart/Recovery At restart, TM reading the log drives RM recovery. Single log scan. Single resolver of transactions. Multiple logs possible, but more complex/more work. Gray & Reuter: Resource Manager 11

End of Whirl-Wind Tour Gray & Reuter: Resource Manager 12

Resource Manager Concepts: Undo Redo Protocol Gray & Reuter: Resource Manager 13

Resource Manager Concepts: Transaction UNDO Protocol declare cursor for transaction_log select rmid, lsn /* a cursor on the transaction's log */ from log /* it returns the resource manager name */ where trid = : trid /* and record id (log sequence number) */ descending lsn; /* and returns records in LIFO order */ void transaction_undo(TRID trid) /* Undo the specified transaction. */ { int sqlcode; /* event variables set by sql */ open cursor transaction_log; /* open an sql cursor on the trans log */ while (TRUE) /* scan trans log backwards & undo each*/ { /* fetch the next most recent log rec */ fetch transaction_log into : rmid, : lsn; /* */ if (sqlcode != 0) break; /* if no more, trans is undone, end loop */ rmid. undo(lsn); /* tell RM to undo that record */ } /* tell RM to undo that record */ close cursor transaction_log; /* Undo scan is complete, close cursor */ }; /* return to caller */ • If UNDO to savepoint , the UNDO stops at desired savepoint Gray & Reuter: Resource Manager 14

Resource Manager Concepts: Restart REDO Protocol void log_redo(void) {declare cursor for the_log select rmid, lsn from log ascending lsn; open cursor the_log; while (TRUE) { fetch the_log into : rmid, : lsn; if (sqlcode != 0) break; rmid. redo(lsn); } close cursor the_log; }; /* /* declare cursor from log start forward /* gets RM id and log record id (lsn) /* of all log records. /* in FIFO order /* open an sql cursor on the log table /* Scan log forward& redo each record. /* fetch the next log record /* if no more, then all redone, end loop /* tell RM to redo that record /* Redo scan complete, close cursor /* return to caller */ */ */ Note: REDO forwards, UNDO backwards Gray & Reuter: Resource Manager 15

Idempotence Old State undo log record New State redo log record F(F(X)) == F(X): Needed in case restart fails (and restarts) Redo(old_state, log) = Redo(new_state, log) = new_state Undo(new_state, log) = Undo(old_state, log) = old_state Gray & Reuter: Resource Manager 16

Testable State: Can Tell If It Happened. IF operation not idempotent AND state not testable THEN recovery is impossible ELSE for F in {UNDO, REDO}: not testable: WHILE (! ACK) F(F(X)) testable: WHILE ( not desired state) {F(x)} Gray & Reuter: Resource Manager 17

Real Operations: Can Not Be Undone Defer operations until commit is assured. Perform as part of Phase 2 of commit If must undo for some reason, generate compensation log record to be processed by some higher authority. Gray & Reuter: Resource Manager 18

Example: Communications Session RM Ops are idempotent (sequence numbers) and testable (sequence numbers) Gray & Reuter: Resource Manager 19

Kinds of Logging Physical: Keep old and new value of container (page, file, . . . ) Pro: Simple Allows recovery of physical object (e. g. broken page) Con: Generates LOTS of log data Logical: -1 Keep call params such that you can compute F(x), F (x) Pro: Sounds simple Compact log. Con: Doesn't work (wrong failure model). Operations do not fail cleanly. Gray & Reuter: Resource Manager 20

Sample Physical LOG RECORD struct compressed_log_record_for_page_update /* */ { int opcode; /* opcode will say compressed page update*/ filename fname; /* name of file that was updated */ long pageno; /* page that was updated */ long offset; /* offset within page that was updated */ long length; /* length of field that was updated */ char old_value[length]; /* old value of field */ char new_value[length]; /* new value of field */ }; /* */ Ordinary sequential insert is OK. Update of sorted (B-tree) page: update LSN update page space map update pointer to record insert record at correct spot (move 1/2 the others) Essentially writes whole page (old and new). 16 KB log records for 100 -byte updates. Gray & Reuter: Resource Manager 21

Sample Physical LOG RECORD struct logical_log_record_for_insert { int opcode; filename fname; long length; char record[length]; }; /* /* opcode will says insert /* name of file that was updated /* length of record that was updated /* value record /* */ */ */ Very compact. Implies page update(s) for record (may be many pages long). Implies index updates (many be many indices on base table) Gray & Reuter: Resource Manager 22

The trouble with Logical Logging Logical logging needs to start UNDO/REDO with an action-consistent state. No half completed operations. for example: insert (table, record) ALL or NONE of the indices should be updated when logical UNDO/REDO is invoked. Problem: Failure model is Page & Message action consistency (Lampson /Sturgis model of Chapter 3). Actions can fail due to: Logic: e. g. duplicate key. Limit: ran out of space Contention: deadlock Media: broken page or session System: computer failure/restart Gray & Reuter: Resource Manager 23

Making Logical Logging Work: Shadows Keep old copy of each page Reset page to old copy at abort (no undo log) Discard old copy at commit. Handles all online failures due to: Logic: e. g. duplicate key. Limit: ran out of space Contention: deadlock Problem: forces page locking, only one updater page. What about restart? Need to atomically write out all changed pages. Gray & Reuter: Resource Manager 24

Making Logical Logging Work: Shadows Perform same shadow trick at disc level. Keep shadow copy of old pages. Write out new pages. In one careful write, write out new page root. Makes update atomic Gray & Reuter: Resource Manager 25

Shadows Pro: Simple Not such a bad deal with non-volatile ram Con: page locking extra space extra overhead (for page maps) extra IO declusters sequential data Gray & Reuter: Resource Manager 26

Compromise Physio-Logical Logging Physical to a "page" (physical container) Logical within a "page". Keep old and new value of container (page, file, . . . ) Pro: Simple Allows recovery of physical object (e. g. broken page) Con: Generates LOTS of log data Gray & Reuter: Resource Manager 27

Logical vs Physio-logical Logging Note: physical log records would be bigger for sorted pages. Gray & Reuter: Resource Manager 28

Physiological Logging Rules Complex operations are a sequence of simple operations on pages and messages. Each operation is constructed as a mini-transaction: lock the object in exclusive mode transform the object generate an UNDO-REDO log record log LSN in object unlock the object. Action Consistent Object: When object semaphore free, no ops in progress. Log-Consistency: contains log records of all complete page/msg actions. Gray & Reuter: Resource Manager 29

Physiological Logging Rules Online Operation - Only Need the Fix Rule Each operation is structured as a mini-transaction. Each operation generates an UNDO record. No page operation fails with the semaphore set. (exception handler must clean up state and UNFIX any pages). Then Rollback can be physical to a page/session/container and logical within page/session/container. Gray & Reuter: Resource Manager 30

Physiological Logging Rules Restart Operation - Need WAL and F@C Need Page-Action consistent disc state. Pages are action consistent. Committed actions can be redone from log. Uncommitted actions can be undone from log. WAL: Write Ahead Log Write undo/redo log records before overwriting disc page Only write action-consistent pages Force-Log-At-Commit Make transaction log records durable at commit. Gray & Reuter: Resource Manager 31

Physiological Logging Rules WAL and F@C WAL: Write Ahead Log write page: get page semaphore copy page give page semaphore /* avoids holding semaphore during IO */ Force_log(Page(LSN)) /*WAL logic, probably already flushed*/ Write copy to disc. WAL gives idempotence and testability. Force-Log-At-Commit At commit phase 1: Force_log(transaction. max_lsn) Gray & Reuter: Resource Manager 32

WAL & F@C in Pictures Volatile Page Versions Volatile Log Durable Log Persistent Page Records Versions PVlsn VLlsn T e m i DLlsn online: VVlsn = VLlsn restart: DLlsn <= VVlsn PVlsn <= DLlsn Commit: commit_lsn <= DLlsn At restart all volatile memory is reset and must be reconstructed from persistent memory. DLlsn restart: PVlsn <= DLlsn commit_lsn <= DLlsn FIX, WAL and F@C assure these assertions Gray & Reuter: Resource Manager 33

The One Bit Resource Manager Manages an array of transactional bits (the free space bit map). i = get_bit(); /* gets a free bit and sets it */ give_bit(i); /* returns a free bit (when transaction commits) */ Gray & Reuter: Resource Manager 34

The Bitmap and Its Log Records The Data Structure struct { LSN xsemaphore Boolean } page; The Log Records struct { int index; Boolean } log_rec; /* layout of the one-bit RM data structure */ lsn; /* page LSN for WAL protocol */ sem; /* semaphore regulates access to the page */ bit[BITS]; /* page. bit[i] = TRUE => bit[i] is free */ /* allocates the page structure */ value; /* log record format for the one-bit RM */ /* index of bit that was updated */ /* new value of bit[index] */ /* log record used by the one-bit RM */ const int rec_size = sizeof(log_rec); /*size of the log record body. Gray & Reuter: Resource Manager */ 35

Page and Log Consistency for 1 -Bit RM Data dirty if reflects an uncommitted transaction update Otherwise, data is clean. Page Consistency: • No clean free bit has been given to any transaction. • Every clean busy bit was given to exactly one transaction. • Dirty bits locked in X mode by updating transactions. • The page. lsn reflects most recent log record for page. Log Consistency: • Log contains a record for every completed mini-transaction update to the page. Gray & Reuter: Resource Manager 36

give_bit() get_bit() & give_bit(i) temporarily violate page consistency. Mini-transaction holds semaphore while violating consistency. Makes page & log mutually consistent before releasing sem. => each mini-transaction observes a consistent page state. void give_bit(int i) /* free a bit { if (LOCK_GRANTED==lock(i, LOCK_X, LOCK_LONG, 0)) /* Lock bit { Xsem_get(&page. sem); /* get page sem page. bit[i] = TRUE; /* free the bit log_rec. index = i; /* generate log rec log_rec. value = TRUE; /*saying bit is free page. lsn = log_insert(log_rec, rec_size); /*write log rec&update lsn Xsem_give(&page. sem); } /* page consistent else /* if lock failed, caller doesn't own bit, Abort_Work(); /* in that case abort caller's trans return; }; /* Gray & Reuter: Resource Manager */ */ */ 37

get_bit() int get_bit(void) /* allocate a bit to and returns bit index */ { int i; /* loop variable */ Xsem_get(&page. sem); /* get the page semaphore */ for ( i = 0; i<BITS; i++); /* loop looking for a free bit */ {if (page. bit[i]) /* if bit is free, may be dirty (so locked)*/ {if (LOCK_GRANTED =lock(i, LOCK_X, LOCK_LONG, 0)); /* lock bit */ { page. bit[i] =FALSE; /* got lock on it, so it was free */ log_rec. value = FALSE; /* generate log rec describing update */ log_rec. index = i; /* */ page. lsn = log_insert(log_rec, rec_size); /* write log rec&update lsn */ Xsem_give(&page. sem); /* page now consistent, give up sem */ return i; } /* return to caller */ }; /* else lock bounce so bit dirty */ }; /* try next free bit, */ Xsem_give(&page. sem); /* if no free bits, give up semaphore */ Abort_Work(); /* abort transaction */ return -1; }; /* returns -1 if no bits are available. */ Gray & Reuter: Resource Manager 38

Compensation Logging Undo may generate a log recording undo step Makes Page LSN monotonic Similar technique was used for Communication Manager (session sequence number was monotonic) Gray & Reuter: Resource Manager 39

1 -bit RM UNDO Callback void undo(LSN lsn) /* undo a one-bit RM operation */ { int i; /* bit index */ Boolean value; /* old bit value from log rec to be undone*/ log_rec_header; /* buffer to hold log record header */ rec_size = log_read_lsn(lsn, header, 0, log_rec, big); /* read log rec */ Xsem_get(&page. sem); /* get the page semaphore */ i = log_rec. index; /* get bit index from log record */ value = ! log_rec. value; /* get complement of new bit value */ page. bit[i] = value; /* update bit to old value */ log_rec. value= value; /* make a compensation log record */ page. lsn = log_insert(log_rec, rec_size); /* log it and bump page lsn */ Xsem_give(&page. sem); /* free the page semaphore */ return; } /* */ Gray & Reuter: Resource Manager 40

1 -bit RM Checkpoint Callback LSN checkpoint(LSN * low_water) /* copy 1 -page RM state to persistent store*/ { Xsem_get(&page. sem); /* get the page semaphore */ *low_water = log_flush(page. lsn); /* WAL force up to page lsn, and */ /* set low water mark */ write(file, page, 0, sizeof(page)); /* write page to persistent memory */ Xsem_give(&page. sem); /* give page semaphore */ return NULLlsn; } /* return checkpoint lsn (none needed) */ Gray & Reuter: Resource Manager 41

1 -bit RM REDO Callback void redo( LSN lsn) /* redo an free space operation */ { int i; /* bit index */ Boolean value; /* new bit value from log rec to be redone*/ log_rec_header; /* buffer to hold log record header */ rec_size = log_read_lsn(lsn, header, 0, log_rec, big); /* read log record */ i = log_rec. index; /* Get bit index */ lock(i, LOCK_X, LOCK_LONG, 0); /* get lock on the bit (often not needed) */ Xsem_get(&page. sem); /* get the page semaphore */ if (page. lsn < lsn) /* if bit version older than log record */ { value= log_rec. value; /* then redo the op. get new bit value */ page. bit[i] = value; /* apply new bit value to bit */ page. lsn = lsn; } /* advance the page lsn */ Xsem_give(&page. sem); /* free the page semaphore */ return; }; /* */ Gray & Reuter: Resource Manager 42

1 -BIT Rm Noise Callbacks Boolean prepare(LSN * lsn) {*lsn = NULLlsn; return TRUE ; }; /* 1 -bit RM has no phase 1 work /* */ */ void Commit(void ) /* Commit release locks & { unlock_class(LOCK_LONG, TRUE, My. RMID()); }; /* return */ */ void Abort(void ) /* Abort release all locks & { unlock_class(LOCK_LONG, TRUE, My. RMID()); }; /* return */ */ Boolean savepoint((LSN * lsn) {*lsn = NULLlsn; return TRUE ; }; */ */ /* no work to do at savepoint /* void UNDO_savepoint(LSN lsn) /* rollback work or abort transaction {if (savepoint == 0) /* if at savepoint zero (abort) unlock_class(LOCK_LONG, TRUE, My. RMID()); /* release all locks }; /* Gray & Reuter: Resource Manager */ */ 43

Summary Model: Complex actions are a page/message action sequence. LSN: Each page carries an LSN and a semaphore. Read. Fix: Read acts semaphore in shared mode. Write. Fix: Update actions get semaphore in exclusive mode, generate one or more log records covering the page, advance the page LSN to match highest LSN give semaphore WAL: log_flush(page. LSN) before overwriting persistent page F@C: force all log records up to the commit LSN at commit Compensation Logging: Invalidate undone log record with a compensating log record. Idempotence via LSN: page LSN makes REDO idempotent Gray & Reuter: Resource Manager 44

Two Phase Commit Getting two or more logs to agree Getting two or more RMs to agree Atomically and Durably Even in case one of them fails and restarts. The TM phases Prepare. Invoke each joined RM asking for its vote. Decide. If all vote yes, durably write commit log record. Commit. Invoke each joined RM, telling it commit decision. Complete. Write commit completion when all RM ACK. Gray & Reuter: Resource Manager 45

Centralized Case of Two Phase Commit Each participant: (TM &RM) goes through a sequence of states Null Active Prepared Committing Committed Aborting Aborted These generate log records Gray & Reuter: Resource Manager 46

Examples Committed Aborted begin DO rm 1 DO rm 2 prepare rm 2 {locks} UNDO rm 2 commit { rm 1, rm 2} UNDO rm 2 complete UNDO rm 1 UNDO begin { rm 1, rm 2} complete Gray & Reuter: Resource Manager 47

Transitions in Case of Restart Active state not persistent, others are persistent For both TM and RM. Log records make them persistent (redo) TM tries to drive states to the right. (to committed, aborted) Gray & Reuter: Resource Manager 48

Successful two phase commit Message/Call flow from TM to each RM joined to transaction If TM and RM share the same log, the RM FORCE can piggyback on the TM FORCE One IO to commit a transaction (less if commit is grouped) Gray & Reuter: Resource Manager 49

Abort Two Phase Commit If RM sends "NO" or no response (timeout), TM starts abort. Calls UNDO of each trans log record May stop at a savepoint. At begin_trans it calls ABORT() callback of each joined RM Gray & Reuter: Resource Manager 50

Distributed two phase commit Tracking joined TMs -- the communications manager helps Much as TRPC helps in the local case. Root TM owes a Prepare/Commit/Abort message to each joined TM. Joined TM does "local" commit. Gray & Reuter: Resource Manager 51

Full Transaction State Diagram Next section explains how these states are implemented. Gray & Reuter: Resource Manager 52

Summary of Resource Manager Concepts DO/UNDO/REDO Idempotent, Testable, Real operations Logical vs Physical logging Shadows to make logical logging work Physiological logging Fix, WAL, Force-at-commit Page/Message/Log consistency RM callbacks (the 1 -bit resource manager) Join, Prepare, Commit, Abort, UNDO, REDO, . . Restart REDO/UNDO Two phase commit (RM story is simple). Gray & Reuter: Resource Manager 53