Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
MongoDB Transaction Pattern
Search
Ian Yang
May 20, 2013
Technology
0
390
MongoDB Transaction Pattern
Ian Yang
May 20, 2013
Tweet
Share
More Decks by Ian Yang
See All by Ian Yang
Migrate to JRuby
doitian
4
770
Tmux Cheatsheet
doitian
2
300
Javascript Test Frameworks
doitian
1
140
Backbone Intro
doitian
1
230
A Tour of Go
doitian
4
880
Other Decks in Technology
See All in Technology
なぜSaaSがMCPサーバーをサービス提供するのか?
sansantech
PRO
8
2.7k
おやつは300円まで!の最適化を模索してみた
techtekt
PRO
0
290
2025年になってもまだMySQLが好き
yoku0825
8
4.5k
なぜテストマネージャの視点が 必要なのか? 〜 一歩先へ進むために 〜
moritamasami
0
200
機械学習を扱うプラットフォーム開発と運用事例
lycorptech_jp
PRO
0
220
Webアプリケーションにオブザーバビリティを実装するRust入門ガイド
nwiizo
5
690
Skrub: machine-learning with dataframes
gaelvaroquaux
0
120
エラーとアクセシビリティ
schktjm
1
1.2k
自作JSエンジンに推しプロポーザルを実装したい!
sajikix
1
170
テストを軸にした生き残り術
kworkdev
PRO
0
190
BPaaSにおける人と協働する前提のAIエージェント-AWS登壇資料
kentarofujii
0
130
「どこから読む?」コードとカルチャーに最速で馴染むための実践ガイド
zozotech
PRO
0
250
Featured
See All Featured
A designer walks into a library…
pauljervisheath
207
24k
How GitHub (no longer) Works
holman
315
140k
Let's Do A Bunch of Simple Stuff to Make Websites Faster
chriscoyier
507
140k
Docker and Python
trallard
45
3.6k
Fireside Chat
paigeccino
39
3.6k
The Language of Interfaces
destraynor
161
25k
Optimizing for Happiness
mojombo
379
70k
Reflections from 52 weeks, 52 projects
jeffersonlam
352
21k
How To Stay Up To Date on Web Technology
chriscoyier
790
250k
Facilitating Awesome Meetings
lara
55
6.5k
Practical Orchestrator
shlominoach
190
11k
Building Flexible Design Systems
yeseniaperezcruz
328
39k
Transcript
MongoDB Transaction Pattern Who? Ian Yang From? Intridea Inc. When?
2013-05-15 Wed
Outline 1 Overview 2 Write Operation Properties 3 Two Phrase
Commit 4 Recovering 5 Rollback 6 Multiple Applications 7 References
Transaction (ACID) Atomicity all or nothing. Consistency does not violate
any integrity constraints Isolation allow concurrent execution of transactions Durability Transaction is persistent after commit
MongoDB Features Atomicity Only on the level of a single
document. Write Concern Return write operation result
MongoDB Transaction Atomicity done or rollback Consistency rollback may leave
system in inconsistent state. Isolation other transactions can see intermediate result. Durability write concern, replica set, journal.
Requirements of Write Operations in Transaction Revertible for recover and
rollback Idempotent for resume, recover and rollback
Not Revertible db.users.save(user); db.users.update ({ username: "ian"}, user); db.users.update ({
username: "ian"}, {$set: {balance: 120}});
Revertible db.users.update ({ username: "ian"}, {$inc: {balance: 20}}); db.users.update ({
username: "ian"}, {$inc: {balance: -20}}); db.users.update ({ username: "ian"}, {$push: {tx: 20}}); db.users.update ({ username: "ian"}, {$pull: {tx: 20}});
Not Idempotent db.users.update ({ username: "ian"}, {$inc: {balance: 20}}); db.users.update
({ username: "ian"}, {$push: {tx: 20}}); db.users.update( {username: "ian", balance: 100}, {$inc: {balance: 20}} );
Idempotent db.users.update( {username: "ian", tx: {$ne: tx._id}}, {$inc: {balance: 20},
$push: {tx: tx._id}} ); db.users.update( {username: "ian", tx: tx._id}, {$inc: {balance: -20}, $pull: {tx: tx._id}} );
Sample Transfer 20 dollars from Ian to Daniel
Create Transaction Transaction provides a unique identifier, enough information to
recover and rollback, a document for atomic write operation.
var tx = { _id: ObjectId (), source: ian._id ,
dest: daniel._id , amount: 20 }; db.transactions.create(tx);
Transaction State Machine State represents transaction stages Different actions should
be taken in different stages to recover or rollback
Step 1: Create Transaction var tx = { _id: ObjectId
(), source: ian._id , dest: daniel._id , amount: 20, state: 'initial ' }; db.transactions.create(tx);
Step 2: initial → pending db.transactions.update( {_id: tx._id , state:
'initial '}, {$set: {state: 'pending '}} );
Step 3: Apply Transaction db.users.update( {_id: tx.source , tx: {$ne:
tx._id}}, {$push: {tx: tx._id}, $inc: { balance: - tx.amount }} ); db.users.update( {_id: tx.dest , tx: {$ne: tx._id}}, {$push: {tx: tx._id}, $inc: { balance: tx.amount }} );
Step 4: pending → committed db.transactions.update( {_id: tx._id , state:
'pending '}, {$set: {state: 'committed '}} );
Step 5: Remove Pending Transaction db.users.update( {_id: tx.source , tx:
tx._id}, {$pull: {tx: tx._id}} ); db.users.update( {_id: tx.dest , tx: tx._id}, {$pull: {tx: tx._id}} );
Step 6: committed → done db.transactions.update( {_id: tx._id , state:
'committed '}, {$set: {state: 'done'}} );
Recover Initial Transactions Recover from Step 2: initial → pending
Recover Pending Transactions Recover from Step 3: Apply Transactions
Recover Committed Transactions Recover from Step 5: Remove Pending Transaction
General Rule Recover from the Step just following the step
transaction just entered current state. State transition creates a check point
Rollback Operations Better to create a new transaction in reverse
direction for Committed Transactions and Done Transactions Use following steps to rollback initial and pending transactions
Rollback 1: state → canceling db.transactions.update( {_id: t._id , state:
{$in: ['initial ', ' pending ']}}, {$set: {state: 'canceling '}} );
Rollback 2: Undo the Transaction db.users.update( {_id: tx.dest , tx:
{$ne: tx._id}}, {$pull: {tx: tx._id}, $inc: { balance: - tx.amount }} ); db.users.update( {_id: tx.source , tx: tx._id}, {$pull: {tx: tx._id}, $inc: { balance: tx.amount }} );
Rollback 3: canceling → canceled db.transactions.update( {_id: t._id , state:
{$in: ['canceling ']}}, {$set: {state: 'canceled '}} );
Recover Rollback Restart from Rollback 2 for canceling transactions
Conclusion Create a document for any transaction Store transaction info
in the document Apply only Revertible and Idempotent operations
Conclusion (Cont.) Transit state to save current work Transit state
if several operation branches exists
Enough? Only 1 application can handle a give transaction at
any point in time.
Background A picks a pending transaction to recover B picks
also a pending transaction to recover
Scenario 1 A is so slow, when it starts apply
transaction, B already has done the transaction. A must rollback. If A exists abnormally, the transaction is in inconsistent state.
Scenario 2 B just finished Step 5: Removing Pending Transaction
A applied transactions and exited abnormally B change state to done without error
Solution findAndModify lock db.transactions.findAndModify ({ query: {state: "initial", application: {$exists:
0}}, update: {$set: {state: "pending", application: "A1"}}, new: true });
Lock Release db.transactions.findAndModify ({ query: {application: "A1"}, update: {$unset: {application:
1}}, new: true });
Lock Forever? Application exists abnormally before release the lock Solution:
timeout
Timeout Lock now = new Date(); timeout = new Date(now.valueOf
() + 60 * 10 * 1000); db.transactions.findAndModify ({ query: {state: "initial", lockUntil: {$not: {$gte: now}}}, update: {$set: {state: "pending", lockUntil: timeout}}, new: true });
Client Requirements Client timeout to ensure the operation is finished
in expected time. Check time after return from block IO Defer transactions cleanup
References http://docs.mongodb.org/ manual/tutorial/ perform-two-phase-commits/ http://docs.mongodb.org/ manual/tutorial/ isolate-sequence-of-operations/
Q & A