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
大規模になったサービスでやるべき基本的なこと
Search
Isao Shimizu
February 07, 2017
Technology
6
2.9k
大規模になったサービスでやるべき基本的なこと
2017年2月7日
【DMM GAMES主催!】「複雑・大規模webサービスを支える技術勉強会」
Isao Shimizu
February 07, 2017
Tweet
Share
More Decks by Isao Shimizu
See All by Isao Shimizu
Notion x ポストモーテムで広げる組織の学び / Notion x Postmortem
isaoshimizu
1
280
New Relicを活用したSREの最初のステップ / NRUG OKINAWA VOL.3
isaoshimizu
3
1k
「家族アルバム みてね」における運用管理・ オブザーバビリティの全貌 / Overview of Operation Management and Observability in FamilyAlbum
isaoshimizu
5
2.2k
約10年間MIXIのインフラを 支えてきたPagerDutyの活用事例 / PagerDuty on Tour 2024
isaoshimizu
6
1.2k
家族アルバム みてねにおけるGrafana活用術 / Grafana Meetup Japan Vol.1 LT
isaoshimizu
2
1.8k
家族アルバム みてねで直面してきた技術的負債 / MIXI KAG 2024
isaoshimizu
18
9.1k
今年1年のEKS運用振り返り/3-shake SRE Tech Talk
isaoshimizu
2
400
ポストモーテムの基礎知識と最新事例 / Fundamentals of Postmortem
isaoshimizu
12
3.2k
全世界1,800万人が利用する「家族アルバム みてね」におけるNew Relic活用法 / FutureStack Tokyo 2023
isaoshimizu
1
600
Other Decks in Technology
See All in Technology
「改善」ってこれでいいんだっけ?
ukigmo_hiro
0
390
Kubernetes self-healing of your workload
hwchiu
0
320
映像エッジAIにおけるNode-RED活用事例
emirmatsui
0
120
Introdução a Service Mesh usando o Istio
aeciopires
1
270
Dylib Hijacking on macOS: Dead or Alive?
patrickwardle
0
430
Click A, Buy B: Rethinking Conversion Attribution in ECommerce Recommendations
lycorptech_jp
PRO
0
120
OAuthからOIDCへ ― 認可の仕組みが認証に拡張されるまで
yamatai1212
0
150
Implementing and Evaluating a High-Level Language with WasmGC and the Wasm Component Model: Scala’s Case
tanishiking
0
160
Introduction to Bill One Development Engineer
sansan33
PRO
0
300
Digitization部 紹介資料
sansan33
PRO
1
5.6k
Findy Team+ QAチーム これからのチャレンジ!
findy_eventslides
0
490
現場データから見える、開発生産性の変化コード生成AI導入・運用のリアル〜 / Changes in Development Productivity and Operational Challenges Following the Introduction of Code Generation AI
nttcom
1
420
Featured
See All Featured
The Myth of the Modular Monolith - Day 2 Keynote - Rails World 2024
eileencodes
26
3.1k
Optimising Largest Contentful Paint
csswizardry
37
3.5k
How STYLIGHT went responsive
nonsquared
100
5.8k
Cheating the UX When There Is Nothing More to Optimize - PixelPioneers
stephaniewalter
285
14k
Measuring & Analyzing Core Web Vitals
bluesmoon
9
630
Principles of Awesome APIs and How to Build Them.
keavy
127
17k
Bash Introduction
62gerente
615
210k
Making Projects Easy
brettharned
120
6.4k
The World Runs on Bad Software
bkeepers
PRO
72
11k
The Pragmatic Product Professional
lauravandoore
36
7k
Let's Do A Bunch of Simple Stuff to Make Websites Faster
chriscoyier
508
140k
Fight the Zombie Pattern Library - RWD Summit 2016
marcelosomers
234
17k
Transcript
େنʹͳͬͨαʔϏεͰΔ͖جຊతͳ͜ͱ ήʔϜ։ൃ෦ SREάϧʔϓ ਗ਼ਫ ܄ @isaoshimizu 2017.2.7 ʲDMM GAMESओ࠵ʂʳʮෳࡶɾେنwebαʔϏεΛࢧ͑Δٕज़ษڧձʯ XFLAG
STUDIO
About me • ਗ਼ਫ ܄ @isaoshimizu • 2011.8-2014.3 SNS mixiͷӡ༻
• XFLAG ελδΦ • 2014.4-2016.6 αʔόʔΤϯδχΞ • 2016.7- SRE ओʹࠃ͚ϞϯελʔετϥΠΫΛࢧ͑Δ͓ࣄ • ଞʹϞϯετελδΞϜɺϒϥφΠDASHͳͲ • ͖ͳͷ • LinuxɺMySQLɺnginxɺMemcachedͳͲͷϛυϧΣΞɺGoɺΫϥϑτϏʔϧ 2
ۙͷൃදࢿྉ • 2016.3.1 ϞϯετΛࢧ͑ΔΠϯϑϥͷࠓͱ͜Ε͔Β https://speakerdeck.com/isaoshimizu/monsutowozhi-eruinhurafalsejin- tokorekara • 2017.1.30 SREάϧʔϓ͕Ͱ͖ͯ͜ͷ͖ؒͬͯͨ͜ͱ https://speakerdeck.com/isaoshimizu/sregurupugadekitekofalseban-nian-
jian-yatutekitakoto 3
αʔϏεͷن͕େ͖͘ͳΔͱ͍͏͜ͱ
αʔϏεͷن͕େ͖͘ͳΔͱ͍͏͜ͱ • ΞΫηε͕૿͑Δ • ෛՙͱͷઓ͍͕࢝·Δ • αʔό͕૿͑Δ • ؔΘΔਓ૿͑Δ 5
ෛՙΛՄࢹԽ͢Δ
ࢹɾϞχλϦϯά • ʮͳΜͱͳ͘ෛՙߴͦ͏ʯͱ͍͏ঢ়ଶͱͯةݥ • ศརͳπʔϧ͕૿͖͍͑ͯͯΔͷͰͪΌΜͱ׆༻͢Δ • Nagios, CloudForecast, Cacti, Ganglia,
Munin, Sensu, Zabbix, Kurado, Prometheus • Mackerel, Datadog, CloudWatch, NewRelic, PagerDuty, Pingdom • Fluentd+Elasticsearch+KibanaͰϩά͔ΒάϥϑΛੜ͢ΔͷޮՌత • άϥϑຖ֬ೝ͢ΔɻཧղͰ͖ͳ͍άϥϑੜɾදࣔͯ͠ҙຯ͕ͳ͍ɻ • ຖ݄ෛՙঢ়گΛαϚϥΠζ͓ͯ͘͠ͱৼΓฦΓࠓޙͷରࡦํ͕࡞Γ͍͢ • ݟ͑ͳ͍ͷɺଌΕͳ͍ͷϦεΫ • αʔϏε࿈ಈͷاըΠϕϯτૣΊʹڞ༗ͯ͠උ͑Δ 7
εέʔϧΞοϓ͔εέʔϧΞτ͖͔͢
εέʔϧΞοϓ or εέʔϧΞτ • εέʔϧΞοϓ • ੑೳΛ্͛ͯෛՙରࡦ • εέʔϧͰ͖Δ্ݶ͕ܾ·ͬͯ͠·͏ɻ͞ΒͳΔෛՙ૿ͷରԠݫ͍͠ɻ •
PCI-ExpressͳioMemoryNVMe SSDΛͬͯμϝͳ࣌μϝɻ • εέʔϧΞτ • LBԼͷΞϓϦέʔγϣϯαʔό૿͢͜ͱͰCPUϦιʔεΛෛՙࢄ • DBͷଓͱΞΫηε͕૿͑ΔͷͰཁҙ • DBʹ͓͍ͯɺεέʔϧΞτͰ͖Δͱɺதظతͳෛՙରࡦͱͯ͠༗ޮ • Ұ࣌తͳෛՙ૿ͷରࡦͰ͋ΕεέʔϧΞοϓޮՌత 9
τϥϑΟοΫͷ૿Ճ
τϥϑΟοΫͷ૿Ճ • Ϋϥυͷ߹ • ΠϯελϯελΠϓʹΑ্ͬͯݶ͕ҟͳΔʢϕϯνϚʔΫͱ͓ͬͯ͘ͱ҆৺ʣ • ࣗࣾDCΛͬͯΔ߹ • NICͱεΠονͷଳҬ্ݶʢαʔόʔ͕ͲͷϥοΫʹ͋Δ͔Λҙࣝʣ •
ճઢͷੑʢํܥ͕མͪͯڐ༰Ͱ͖ΔଳҬઃܭʣ • NICͷੑ • Bonding: ίωΫγϣϯ୯ҐͰࢄ͞ΕΔͷͰNICx2Ͱ2ഒͷଳҬͱݶΒͳ͍ • TCPͷ࠶ૹճͷࢹ֮Խ • ಛʹΫϥυͰϨΠςϯγଳҬͷมԽ͕͋ΔͷͰཁҙ 11
αʔόʔ/Πϯελϯεͷ૿Ճ
αʔόʔ/Πϯελϯεͷ૿Ճ • σϓϩΠɺϓϩϏδϣχϯάͷޮԽɺࣗಈԽ • Capistrano, AWS CodeDeploy, Stretcher + Consul
• Chef, Ansible, Puppet, Itamae • ϕʔεͱͳΔΠϝʔδʢAMIͳͲʣߏஙͷࣗಈԽ • Packerศར • ʹґଘ͠ͳ͍ΦϖϨʔγϣϯ͕ཧܥ 13
σʔλϕʔε ΫΤϦͷ૿Ճ
σʔλϕʔε ΫΤϦͷ૿Ճ • ϘτϧωοΫREAD͔WRITE͔ • σʔλϕʔεɺςʔϒϧͷ୯ҐͰαʔόΛׂ • γϟʔσΟϯάʢIDϋογϡͷ༨Ͱׂʣ • ΩϟογϡʢMemcached,
RedisͳͲʣͷ׆༻ • ετϨʔδͷIOݶքͷݟۃΊʢΧλϩάεϖοΫΛ৴༻ͤͣܭଌ ͢Δʣ • tmp table, file sortͳͲ͕ൃੜ͍ͯ͠ͳ͍͔ • Dirty PagewaitͷൃੜසΛάϥϑ͔ΒΔ • όοϑΝϓʔϧར༻ͷঢ়گ 15
σʔλϕʔε αΠζͷ૿Ճ
σʔλϕʔε αΠζͷ૿Ճ • ҰᷓΕͯ͠·͏ͱରॲ͕ࠔ • ςʔϒϧͷσʔλɺΠϯσοΫεαΠζ • όοϑΝʔϓʔϧͷΩϟύγςΟ • ϝϞϦΛ૿ͯ͠ϓʔϧαΠζΛ͛Δͱ͍͏ख
• όΠφϦϩάͷ૿Ճɻexpire_logs_daysͷௐʢMySQLͷ߹ʣɻ • AUTO INCREMENTͷ্ݶʹҙʢint21ԯɺbigint922ژʣ • intͩͱςʔϒϧʹΑͬͯ͋ͬͱ͍͏ؒʹୡͯ͠͠·͏ͷͰҙ • ᮢΛઃఆͯ͠ࢹ͓ͯ͘͠ͱ҆৺ 17
ਓһͷ૿Ճ
ਓһͷ૿Ճ • ΞΧϯτ • ΫϥυͷΞΧϯτɺLinuxϢʔβʔΞΧϯτɺGitHubͷΞΧϯτ • AWSͷ߹ɺCloudTrailΛͬͨࠪϩά • LDAPͳͲͰϩάΠϯϢʔβʔɺϗετͷ੍ݶʢηΩϡϦςΟͷ୲อʣ •
ϓϩδΣΫτ৽ͨʹδϣΠϯˍൈ͚ΔϝϯόʔͷରԠʢνΣοΫϦετͷςϯϓϨʔτʣ • Ξϥʔτ • ൪੍ɺΤεΧϨʔγϣϯํ๏ͳͲͷཱ֬ɺPagerDutyͷ׆༻ • ϊϋखॱͷڞ༗ • Wikiͷ׆༻ʢྫ: GitHub Wiki, Qiita Team, Crowiʣ • ߏஙোൃੜ࣌ͷखॱɺϧʔϧΛυΩϡϝϯτԽʢΠϯγσϯτൃੜ࣌ʹྫྷ੩ʹରॲͰ͖ΔΑ͏ʹʣ • 4 eyesͰͷ࡞ۀ֬ೝͯ͠ࣄނࢭʢྫ: DNSߋ৽ɺTerminateɺService StopͳͲʣ • ChatOpsͰ࡞ۀͰ͖Δ͜ͱΛ૿͢ 19
·ͱΊ
·ͱΊ • نʹ͔͔ΘΒͣجຊΛ͔ͬ͠Γͱ • ى͖͍ͯΔ͜ͱΛਖ਼͘͠Ѳ͢Δ͜ͱɺΠϝʔδͰ͖Δ͜ͱ • نʹൺྫͯ͠࡞ۀྔ͕૿͑ͳ͍Α͏ʹ͢Δ͜ͱ • ਓ͕૿͑ͨͱ͖ʹ࡞ۀෛՙ͕ࢄ͞ΕΔΑ͏ʹϊϋखॱΛల։͢Δ͜ͱ •
େن͡Όͳͯͬͨ͘΄͏͕͍͍͜ͱଟ͍ • ؒืूத https://xflag.com/recruit/ 21
Thank you!