{"id":35955,"date":"2019-10-31T22:08:26","date_gmt":"2019-10-31T19:08:26","guid":{"rendered":"https:\/\/prohoster.info\/blog\/razvertyvanie-prilozhenij-na-neskolkih-klasterah-kubernetes-s-helm\/"},"modified":"2019-10-31T22:08:26","modified_gmt":"2019-10-31T19:08:26","slug":"razvertyvanie-prilozhenij-na-neskolkih-klasterah-kubernetes-s-helm","status":"publish","type":"post","link":"https:\/\/prohoster.info\/hr\/blog\/administrirovanie\/razvertyvanie-prilozhenij-na-neskolkih-klasterah-kubernetes-s-helm","title":{"rendered":"Implementacija aplikacija na vi\u0161e Kubernetes klastera pomo\u0107u Helma","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"<\/p>\n<p><\/p>\n<h2 id=\"kak-dailymotion-ispolzuet-kubernetes-razvertyvanie-prilozheniy\">Kako Dailymotion koristi Kubernetes: implementacija aplikacija<\/h2>\n<p><\/p>\n<p>U Dailymotionu smo po\u010deli koristiti Kubernetes u produkciji prije 3 godine. No, implementacija aplikacija na vi\u0161e klastera predstavlja izazov, pa smo u posljednjih nekoliko godina nastojali pobolj\u0161ati na\u0161e alate i radne procese.<\/p>\n<p><noindex><a rel=\"nofollow\" name=\"habracut\"><\/a><\/noindex><\/p>\n<h3 id=\"s-chego-nachalos\">Kako je sve po\u010delo<\/h3>\n<p><\/p>\n<p>Ovdje \u0107emo objasniti kako implementiramo na\u0161e aplikacije na vi\u0161e klastera Kubernetes \u0161irom svijeta.<\/p>\n<p><\/p>\n<p>Za implementaciju vi\u0161e Kubernetes objekata odjednom koristimo <noindex><a rel=\"nofollow\" href=\"https:\/\/helm.sh\/\">Helm<\/a><\/noindex>, a svi na\u0161i chartovi pohranjuju se u jednom git repozitoriju. Da bismo implementirali puni softverski paket iz vi\u0161e usluga, koristimo takozvani sveobuhvatni chart. U su\u0161tini, to je chart koji deklarira ovisnosti i omogu\u0107ava inicijalizaciju API-ja i njegovih usluga jednim naredbom.<\/p>\n<p><\/p>\n<p>Tako\u0111er smo napisali mali Python skript preko Helma kako bismo obavljali provjere, kreirali chartove, dodavali tajne i implementirali aplikacije. Sve te zadatke obavljamo na centralnoj CI platformi koriste\u0107i docker sliku.<\/p>\n<p><\/p>\n<p>Pre\u0111imo na stvar.<\/p>\n<p><\/p>\n<blockquote><p>Napomena. Kada ovo \u010ditate, prvi kandidat za izdanja Helma 3 je ve\u0107 objavljen. Glavna verzija sadr\u017ei \u010ditav niz pobolj\u0161anja koja su osmi\u0161ljena da rije\u0161e neke probleme s kojima smo se suo\u010davali u pro\u0161losti.<\/p><\/blockquote>\n<p><\/p>\n<h3 id=\"rabochiy-process-razrabotki-chartov\">Radni proces razvoja chartova<\/h3>\n<p><\/p>\n<p>Za aplikacije koristimo grane, i ovaj pristup smo odlu\u010dili primijeniti i na chartove.<\/p>\n<p><\/p>\n<ul>\n<li>Grana <strong>dev<\/strong> se koristi za izradu chartova koji \u0107e se testirati na razvojnim klasterima.<\/li>\n<li>Kada se pull request po\u0161alje u <strong>master<\/strong>, oni se provjeravaju na stagingu.<\/li>\n<li>Na kraju, kreiramo pull request kako bismo prenijeli promjene u granu <strong>prod<\/strong> i primijenili ih u produkciji.<\/li>\n<\/ul>\n<p><\/p>\n<p>Svako okru\u017eenje ima svoj privatni repozitorij koji pohranjuje na\u0161e chartove, a koristimo <noindex><a rel=\"nofollow\" href=\"https:\/\/chartmuseum.com\/\">Chartmuseum<\/a><\/noindex> s vrlo korisnim API-jima. Tako jam\u010dimo strogu izolaciju izme\u0111u okru\u017eenja i provjeru chartova u stvarnim uvjetima prije nego \u0161to ih upotrijebimo u produkciji.<\/p>\n<p><\/p>\n<p><noindex><a rel=\"nofollow\" href=\"https:\/\/habrastorage.org\/webt\/rz\/vt\/uk\/rzvtuk6q7b_rkeaxnyujzhmfjog.png\"><\/a><\/noindex><\/p>\n<p><\/p>\n<p><em>Repozitoriji chartova u razli\u010ditim okru\u017eenjima<\/em><\/p>\n<p><\/p>\n<p>Va\u017eno je napomenuti da kada developeri \u0161alju granu dev, verzija njihovog charta automatski se \u0161alje u dev Chartmuseum. Tako svi developeri koriste jedan dev repozitorij i treba pa\u017eljivo odrediti svoju verziju charta kako ne bi slu\u010dajno koristili ne\u010dije izmjene.<\/p>\n<p><\/p>\n<p>Osim toga, na\u0161 mali Python skript provjerava Kubernetes objekte prema Kubernetes OpenAPI specifikacijama koriste\u0107i <noindex><a rel=\"nofollow\" href=\"https:\/\/kubeval.instrumenta.dev\/\">Kubeval<\/a><\/noindex>, prije nego \u0161to ih objavimo u Chartmusu.<\/p>\n<p><\/p>\n<h3 id=\"obschee-opisanie-rabochego-processa-razrabotki-charta\">Op\u0107i opis radnog toka razvoja charta<\/h3>\n<p><\/p>\n<p><noindex><a rel=\"nofollow\" href=\"https:\/\/habrastorage.org\/webt\/yd\/cw\/qw\/ydcwqw-vqfitu5bj8ky4nh7xxea.png\"><\/a><\/noindex><\/p>\n<p><\/p>\n<ol>\n<li>Konfiguracija zadataka cjevovoda prema specifikacijama <noindex><a rel=\"nofollow\" href=\"https:\/\/gazr.io\/\">gazr.io<\/a><\/noindex> za kontrolu kvalitete (lint, unit-test).<\/li>\n<li>Slanje docker slike s alatima Python koji raspore\u0111uju na\u0161e aplikacije.<\/li>\n<li>Konfiguracija okru\u017eenja prema nazivu grane.<\/li>\n<li>Provjera yaml datoteka Kubernetes pomo\u0107u Kubevala.<\/li>\n<li>Automatsko pove\u0107anje verzije charta i njegovih roditeljskih chartova (chartova koji ovise o promjenjivom chartu).<\/li>\n<li>Slanje charta u Chartmuseum, koji odgovara njegovom okru\u017eenju<\/li>\n<\/ol>\n<p><\/p>\n<h2 id=\"upravlenie-razlichiyami-v-klasterah\">Upravljanje razlikama u klasterima<\/h2>\n<p><\/p>\n<h3 id=\"federaciya-klasterov\">Federacija klastera<\/h3>\n<p><\/p>\n<p>Bilo je vremena kada smo koristili <noindex><a rel=\"nofollow\" href=\"https:\/\/kubernetes.io\/docs\/concepts\/cluster-administration\/federation\/\">federaciju klastera Kubernetes<\/a><\/noindex>, gdje se mogli objavljivati Kubernetes objekti s jedne API to\u010dke. Ali su se pojavili problemi. Na primjer, neki Kubernetes objekti nisu se mogli kreirati na federacijskoj to\u010dki, pa je bilo te\u0161ko odr\u017eavati objedinjene objekte i druge objekte za pojedina\u010dne klastere.<\/p>\n<p><\/p>\n<p>Da bismo rije\u0161ili problem, po\u010deli smo upravljati klasterima neovisno, \u0161to je zna\u010dajno pojednostavilo proces (koristili smo prvu verziju federacije; u drugoj je ne\u0161to moglo biti promijenjeno).<\/p>\n<p><\/p>\n<h3 id=\"georaspredelennaya-platforma\">Geografski distribuirana platforma<\/h3>\n<p><\/p>\n<p>Sada je na\u0161a platforma distribuirana u 6 regija \u2014 3 lokalno i 3 u oblaku.<\/p>\n<p><\/p>\n<p><noindex><a rel=\"nofollow\" href=\"https:\/\/habrastorage.org\/webt\/jq\/c1\/mr\/jqc1mru36g5byams4wddzducauw.png\"><\/a><\/noindex><br \/>\n<em>Distribuirano raspore\u0111ivanje<\/em><\/p>\n<p><\/p>\n<h3 id=\"globalnye-znacheniya-helm\">Globalne Helm vrijednosti<\/h3>\n<p><\/p>\n<p>4 globalne Helm vrijednosti omogu\u0107uju definiranje razlika me\u0111u klasterima. Za sve na\u0161e chartove postoje minimalne pretpostavljene vrijednosti.<\/p>\n<p><\/p>\n<pre><code class=\"plaintext\">global:\n  cloud: True\n  env: staging\n  region: us-central1\n  clusterName: staging-us-central1<\/code><\/pre>\n<p><\/p>\n<p><em>Globalne vrijednosti<\/em><\/p>\n<p><\/p>\n<p>Te vrijednosti poma\u017eu u odre\u0111ivanju konteksta za na\u0161e aplikacije i koriste se za razli\u010dite zadatke: pra\u0107enje, traganje, evidentiranje, izvr\u0161avanje vanjskih poziva, skaliranje itd.<\/p>\n<p><\/p>\n<ul>\n<li>\u00abcloud\u00bb: imamo hibridnu platformu Kubernetes. Na primjer, na\u0161 API se raspore\u0111uje u GCP zonama i u na\u0161im podatkovnim centrima.<\/li>\n<li>\u00abenv\u00bb: neke vrijednosti mogu se mijenjati za neproduktivna okru\u017eenja. Na primjer, definicije resursa i konfiguracije automatskog skaliranja.<\/li>\n<li>\u00abregion\u00bb: ove informacije poma\u017eu u odre\u0111ivanju lokacije klastera i mogu se koristiti za odre\u0111ivanje najbli\u017eih krajnjih to\u010daka za vanjske usluge.<\/li>\n<li>\u00abclusterName\u00bb: ako i kada \u017eelimo odrediti vrijednost za pojedina\u010dni klaster.<\/li>\n<\/ul>\n<p><\/p>\n<p>Evo konkretnog primjera:<\/p>\n<p><\/p>\n<pre><code class=\"plaintext\">{{\n\/* Vra\u0107a Horizontal Pod Autoscaler replike za GraphQL *\/}}\n{{- define \"graphql.hpaReplicas\" -}}\n{{- if eq .Values.global.env \"prod\" }}\n{{- if eq .Values.global.region \"europe-west1\" }}\nminReplicas: 40\n{{- else }}\nminReplicas: 150\n{{- end }}\nmaxReplicas: 1400\n{{- else }}\nminReplicas: 4\nmaxReplicas: 20\n{{- end }}\n{{- end -}}<\/code><\/pre>\n<p><\/p>\n<p><em>Primjer Helm predlo\u0161ka<\/em><\/p>\n<p><\/p>\n<p>Ova logika je definirana u pomo\u0107nom predlo\u0161ku kako bi se izbjeglo preoptere\u0107enje Kubernetes YAML.<\/p>\n<p><\/p>\n<h3 id=\"obyavlenie-prilozheniya\">Objava aplikacije<\/h3>\n<p><\/p>\n<p>Na\u0161i alati za implementaciju temelje se na nekoliko YAML datoteka. Ispod je primjer kako objavljujemo uslugu i njenu topologiju skaliranja (broj replika) u klasteru.<\/p>\n<p><\/p>\n<pre><code class=\"plaintext\">releases:\n  - foo.world\n\nfoo.world:                # Ime izdanja\n  services:               # Lista dailymotion aplikacija\/projekata\n    foobar:\n      chart_name: foo-foobar\n      repo: git@github.com:dailymotion\/foobar\n      contexts:\n        prod-europe-west1:\n          deployments:\n            - name: foo-bar-baz\n              replicas: 18\n            - name: another-deployment\n              replicas: 3<\/code><\/pre>\n<p><\/p>\n<p><em>Definicija usluge<\/em><\/p>\n<p><\/p>\n<p>Ovo je skica svih koraka koji definiraju na\u0161 radni proces implementacije. Zadnji korak implementira aplikaciju istovremeno na nekoliko radnih klastera.<\/p>\n<p><\/p>\n<p><noindex><a rel=\"nofollow\" href=\"https:\/\/habrastorage.org\/webt\/iu\/qh\/xk\/iuqhxki4nfqoi0mus258j0feylu.png\"><\/a><\/noindex><br \/>\n<em>Koraci implementacije u Jenkinsu<\/em><\/p>\n<p><\/p>\n<h3 id=\"a-sekrety\">A tajne?<\/h3>\n<p><\/p>\n<p>\u0160to se ti\u010de sigurnosti, pratimo sve tajne iz razli\u010ditih izvora i \u010duvamo ih u jedinstvenom skladi\u0161tu <noindex><a rel=\"nofollow\" href=\"https:\/\/www.vaultproject.io\/\">Vault<\/a><\/noindex> u Parizu.<\/p>\n<p><\/p>\n<p>Na\u0161i alati za implementaciju izvla\u010de vrijednosti tajni iz Vault-a i, kada do\u0111e vrijeme za implementaciju, umetnu ih u Helm.<\/p>\n<p><\/p>\n<p>Za to smo definirali mapiranje izme\u0111u tajni u Vault-u i tajni potrebnih na\u0161im aplikacijama:<\/p>\n<p><\/p>\n<pre><code class=\"plaintext\">tajne:                                                                                                                                                                                                        \n     - secret_id: \"stack1-app1-password\"                                                                                                                                                                                  \n       konteksti:                                                                                                                                                                                                   \n         - ime: \"default\"                                                                                                                                                                                         \n           vaultPath: \"\\\/kv\\\/dev\\\/stack1\\\/app1\\\/test\"                                                                                                                                                               \n           vaultKey: \"password\"                                                                                                                                                                                    \n         - ime: \"cluster1\"                                                                                                                                                                           \n           vaultPath: \"\\\/kv\\\/dev\\\/stack1\\\/app1\\\/test\"                                                                                                                                                               \n           vaultKey: \"password\"<\/code><\/pre>\n<p><\/p>\n<ul>\n<li>Odredili smo op\u0107a pravila koja je potrebno slijediti prilikom zapisivanja tajni u Vault.<\/li>\n<li>Ako tajna pripada <strong>odre\u0111enom kontekstu ili klasteru<\/strong>, potrebno je dodati specifi\u010dnu stavku. (U ovom slu\u010daju, kontekst cluster1 ima vlastitu vrijednost za tajnu stack-app1-password).<\/li>\n<li>U suprotnom se koristi vrijednost <strong>po defaultu<\/strong>.<\/li>\n<li>Za svaku stavku na ovom popisu u <strong>tajni Kubernetes<\/strong> ubacuje se par klju\u010d-vrijednost. Stoga je predlo\u017eak tajne u na\u0161im chartovima vrlo jednostavan.<\/li>\n<\/ul>\n<p><\/p>\n<pre><code class=\"plaintext\">apiVersion: v1\ndata:\n{{- range $key,$value := .Values.secrets }}\n  {{ $key }}: {{ $value | b64enc | quote }}\n{{ end }}\nkind: Secret\nmetadata:\n  name: \"{{ .Chart.Name }}\"\n  labels:\n    chartVersion: \"{{ .Chart.Version }}\"\n    tillerVersion: \"{{ .Capabilities.TillerVersion.SemVer }}\"\ntype: Opaque<\/code><\/pre>\n<p><\/p>\n<h2 id=\"problemy-i-ogranicheniya\">Problemi i ograni\u010denja<\/h2>\n<p><\/p>\n<h3 id=\"rabota-s-neskolkimi-repozitoriyami\">Rad s vi\u0161e repozitorija<\/h3>\n<p><\/p>\n<p>Trenutno odvajamo razvoj chartova i aplikacija. To zna\u010di da programeri moraju raditi u dva git repozitorija: jedan za aplikaciju, a drugi za definiciju njenog implementiranja u Kubernetes. 2 git repozitorija predstavljaju 2 radna procesa, i po\u010detnicima lako mo\u017ee biti zbunjuju\u0107e.<\/p>\n<p><\/p>\n<h3 id=\"upravlyat-obobschennymi-chartami-hlopotno\">Upravljanje generi\u010dkim chartovima je zamoran posao<\/h3>\n<p><\/p>\n<p>Kao \u0161to smo ve\u0107 spomenuli, generi\u010dki chartovi su vrlo korisni za definiranje ovisnosti i brzo implementiranje vi\u0161e aplikacija. No, koristimo <code>--reuse-values<\/code>, kako bismo izbjegli prijenos svih vrijednosti svaki put kada implementiramo aplikaciju koja ulazi u ovaj generi\u010dki chart.<\/p>\n<p><\/p>\n<p>U procesu neprekidne isporuke imamo samo dvije vrijednosti koje se redovito mijenjaju: broj replika i oznaku slike (verzija). Ostale, stabilnije vrijednosti, mijenjaju se ru\u010dno, \u0161to je prili\u010dno izazovno. Osim toga, jedna gre\u0161ka u rasporedu generi\u010dkog charta mo\u017ee dovesti do ozbiljnih kvarova, \u0161to smo sami iskusili.<\/p>\n<p><\/p>\n<h3 id=\"obnovlenie-neskolkih-faylov-konfiguracii\">A\u017euriranje vi\u0161e konfiguracijskih datoteka<\/h3>\n<p><\/p>\n<p>Kada programer dodaje novu aplikaciju, mora promijeniti nekoliko datoteka: deklaraciju aplikacije, popis tajni, dodavanje aplikacije u ovisnosti, ako je uklju\u010dena u generi\u010dki chart.<\/p>\n<p><\/p>\n<h3 id=\"razresheniya-jenkins-slishkom-rasshireny-v-vault\">Dozvole Jenkins-a su pre\u0161iroke u Vaultu<\/h3>\n<p><\/p>\n<p>Trenutno imamo jedan <noindex><a rel=\"nofollow\" href=\"https:\/\/www.vaultproject.io\/docs\/auth\/approle.html\">AppRole<\/a><\/noindex>, koji \u010dita sve tajne iz Vaulta.<\/p>\n<p><\/p>\n<h3 id=\"process-otkata-ne-avtomatizirovan\">Postupak vra\u0107anja nije automatiziran<\/h3>\n<p><\/p>\n<p>Za vra\u0107anje je potrebno izvr\u0161iti naredbu na nekoliko klastera, \u0161to mo\u017ee dovesti do gre\u0161aka. Ovu operaciju radimo ru\u010dno kako bismo osigurali da zadajemo ispravni identifikator verzije.<\/p>\n<p><\/p>\n<h2 id=\"my-dvizhemsya-v-storonu-gitops\">Kre\u0107emo se prema GitOps-u<\/h2>\n<p><\/p>\n<h3 id=\"nasha-cel\">Na\u0161 cilj<\/h3>\n<p><\/p>\n<p>\u017delimo vratiti chart u repozitorij aplikacije koju on raspore\u0111uje.<\/p>\n<p><\/p>\n<p>Radni proces \u0107e biti isti kao za razvoj. Na primjer, kada se grana po\u0161alje na master, raspored \u0107e se pokretati automatski. Glavna razlika izme\u0111u ovakvog pristupa i trenutnog radnog procesa \u0107e biti u tome \u0161to <strong>sve \u0107e se upravljati u gitu<\/strong> (sama aplikacija i na\u010din njenog raspore\u0111ivanja u Kubernetes-u).<\/p>\n<p><\/p>\n<p>Postoji nekoliko prednosti:<\/p>\n<p><\/p>\n<ul>\n<li>Mnogo <strong>jasnije<\/strong> za programera. Lak\u0161e je nau\u010diti kako primijeniti promjene u lokalnom chartu.<\/li>\n<li>Definiciju rasporeda usluge mo\u017ee se odrediti <strong>isto tako, gdje je kod<\/strong> usluge.<\/li>\n<li><strong>Upravljanje brisanjem generi\u010dkih chartova<\/strong>. Usluga \u0107e imati svoju Helm verziju. To \u0107e omogu\u0107iti upravljanje \u017eivotnim ciklusom aplikacije (vra\u0107anje, nadogradnja) na najsitnijem nivou, kako ne bi utjecalo na druge usluge.<\/li>\n<li><strong>Prednosti gita<\/strong> za upravljanje chartovima: opoziv promjena, dnevnik revizije itd. Ako je potrebno opozvati promjenu charta, to se mo\u017ee u\u010diniti pomo\u0107u gita. Raspored se pokre\u0107e automatski.<\/li>\n<li>Mo\u017ee se razmisliti o unapre\u0111enju radnog procesa razvoja pomo\u0107u alata kao \u0161to su <strong>Skaffold<\/strong>, s kojima programeri mogu testirati promjene u kontekstu bliskom produkciji.<\/li>\n<\/ul>\n<p><\/p>\n<h3 id=\"dvuhetapnaya-migraciya\">Dvofazna migracija<\/h3>\n<p><\/p>\n<p>Na\u0161i developeri koriste ovaj radni proces ve\u0107 2 godine, stoga nam je potrebna \u0161to bezbolnija migracija. Stoga smo odlu\u010dili dodati intermedijarnu fazu na putu do cilja.<br \/>\nPrva faza je jednostavna:<\/p>\n<p><\/p>\n<ul>\n<li>Zadr\u017eavamo sli\u010dnu strukturu za konfiguraciju implementacije aplikacija, ali u jednom objektu pod nazivom DailymotionRelease.<\/li>\n<\/ul>\n<p><\/p>\n<pre><code class=\"plaintext\">apiVersion: \"v1\"\nkind: \"DailymotionRelease\"\nmetadata:\n  name: \"app1.ns1\"\n  environment: \"dev\"\n  branch: \"mybranch\"\nspec:\n  slack_channel: \"#admin\"\n  chart_name: \"app1\"\n  scaling:\n    - context: \"dev-us-central1-0\"\n      replicas:\n        - name: \"hermes\"\n          count: 2\n    - context: \"dev-europe-west1-0\"\n      replicas:\n        - name: \"app1-deploy\"\n          count: 2\n  secrets:\n    - secret_id: \"app1\"\n      contexts:\n        - name: \"default\"\n          vaultPath: \"\\\/kv\\\/dev\\\/ns1\\\/app1\\\/test\"\n          vaultKey: \"password\"\n        - name: \"dev-europe-west1-0\"\n          vaultPath: \"\\\/kv\\\/dev\\\/ns1\\\/app1\\\/test\"\n          vaultKey: \"password\"<\/code><\/pre>\n<p><\/p>\n<ul>\n<li>1 izdanje po aplikaciji (bez generaliziranih chartova).<\/li>\n<li>Chartovi u git repozitoriju aplikacije.<\/li>\n<\/ul>\n<p><\/p>\n<p>Razgovarali smo sa svim developerima, tako da je proces migracije ve\u0107 po\u010deo. Prva faza se jo\u0161 uvijek kontrolira pomo\u0107u CI platforme. Uskoro \u0107u napisati jo\u0161 jedan post o drugoj fazi: kako smo pre\u0161li na GitOps radni proces s <noindex><a rel=\"nofollow\" href=\"https:\/\/github.com\/weaveworks\/flux\">Flux<\/a><\/noindex>. Ispri\u010dat \u0107u kako smo sve postavili i s kakvim smo se izazovima susreli (nekoliko repozitorija, tajne itd.). Pratite vijesti.<\/p>\n<p><\/p>\n<p>Ovdje smo poku\u0161ali opisati na\u0161 napredak u radnom procesu implementacije aplikacija tijekom posljednjih godina, \u0161to je dovelo do razmi\u0161ljanja o GitOps pristupu. Jo\u0161 nismo postigli cilj i redovito \u0107emo obavje\u0161tavati o rezultatima, ali sada smo uvjereni da smo ispravno postupili kada smo odlu\u010dili sve pojednostaviti i pribli\u017eiti navikama developera.<\/p>\n<p>Izvor: <a content=\"nofollow\" rel=\"nofollow\" href=\"https:\/\/habr.com\/ru\/company\/southbridge\/blog\/458934\/\">habr.com<\/a><\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"<p>\u041a\u0430\u043a Dailymotion \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u0435\u0442 Kubernetes: \u0440\u0430\u0437\u0432\u0435\u0440\u0442\u044b\u0432\u0430\u043d\u0438\u0435 \u043f\u0440\u0438\u043b\u043e\u0436\u0435\u043d\u0438\u0439 \u041c\u044b \u0432 Dailymotion \u043d\u0430\u0447\u0430\u043b\u0438 \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u0442\u044c Kubernetes \u0432 \u043f\u0440\u043e\u0434\u0430\u043a\u0448\u0435\u043d\u0435 3 \u0433\u043e\u0434\u0430 \u043d\u0430\u0437\u0430\u0434. \u041d\u043e \u0440\u0430\u0437\u0432\u0435\u0440\u0442\u044b\u0432\u0430\u0442\u044c \u043f\u0440\u0438\u043b\u043e\u0436\u0435\u043d\u0438\u044f \u043d\u0430 \u043d\u0435\u0441\u043a\u043e\u043b\u044c\u043a\u0438\u0445 \u043a\u043b\u0430\u0441\u0442\u0435\u0440\u0430\u0445 \u0442\u043e \u0435\u0449\u0435 \u0443\u0434\u043e\u0432\u043e\u043b\u044c\u0441\u0442\u0432\u0438\u0435, \u043f\u043e\u044d\u0442\u043e\u043c\u0443 \u0432 \u043f\u043e\u0441\u043b\u0435\u0434\u043d\u0438\u0435 \u043d\u0435\u0441\u043a\u043e\u043b\u044c\u043a\u043e \u043b\u0435\u0442 \u043c\u044b \u0441\u0442\u0430\u0440\u0430\u043b\u0438\u0441\u044c \u0443\u043b\u0443\u0447\u0448\u0438\u0442\u044c \u043d\u0430\u0448\u0438 \u0438\u043d\u0441\u0442\u0440\u0443\u043c\u0435\u043d\u0442\u044b \u0438 \u0440\u0430\u0431\u043e\u0447\u0438\u0435 \u043f\u0440\u043e\u0446\u0435\u0441\u0441\u044b. \u0421 \u0447\u0435\u0433\u043e \u043d\u0430\u0447\u0430\u043b\u043e\u0441\u044c \u0417\u0434\u0435\u0441\u044c \u043c\u044b \u0440\u0430\u0441\u0441\u043a\u0430\u0436\u0435\u043c, \u043a\u0430\u043a \u043c\u044b \u0440\u0430\u0437\u0432\u0435\u0440\u0442\u044b\u0432\u0430\u0435\u043c \u043d\u0430\u0448\u0438 \u043f\u0440\u0438\u043b\u043e\u0436\u0435\u043d\u0438\u044f \u043d\u0430 \u043d\u0435\u0441\u043a\u043e\u043b\u044c\u043a\u0438\u0445 \u043a\u043b\u0430\u0441\u0442\u0435\u0440\u0430\u0445 Kubernetes \u043f\u043e [&hellip;]<\/p>\n","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[688],"tags":[],"class_list":["post-35955","post","type-post","status-publish","format-standard","hentry","category-administrirovanie"],"aioseo_notices":[],"aioseo_head":"\n\t\t<!-- All in One SEO 5.0.4 - aioseo.com -->\n\t<meta name=\"description\" content=\"\u041a\u0430\u043a Dailymotion \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u0435\u0442 Kubernetes: \u0440\u0430\u0437\u0432\u0435\u0440\u0442\u044b\u0432\u0430\u043d\u0438\u0435 \u043f\u0440\u0438\u043b\u043e\u0436\u0435\u043d\u0438\u0439 \u041c\u044b \u0432 Dailymotion \u043d\u0430\u0447\u0430\u043b\u0438 \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u0442\u044c Kubernetes \u0432.\" \/>\n\t<meta name=\"robots\" content=\"max-image-preview:large\" \/>\n\t<meta name=\"author\" content=\"Yuri Gagarin\"\/>\n\t<link rel=\"canonical\" href=\"https:\/\/prohoster.info\/hr\/blog\/administrirovanie\/razvertyvanie-prilozhenij-na-neskolkih-klasterah-kubernetes-s-helm\" \/>\n\t\t<meta name=\"generator\" content=\"All in One SEO (AIOSEO) 5.0.4\" \/>\n\t\t<meta property=\"og:locale\" content=\"hr_HR\" \/>\n\t\t<meta property=\"og:site_name\" content=\"ProHoster | \u041a\u0443\u043f\u0438\u0442\u044c \u043d\u0430\u0434\u0435\u0436\u043d\u044b\u0439 \u0445\u043e\u0441\u0442\u0438\u043d\u0433 \u0434\u043b\u044f \u0441\u0430\u0439\u0442\u043e\u0432 \u0441 \u0437\u0430\u0449\u0438\u0442\u043e\u0439 \u043e\u0442 DDoS, VPS VDS \u0441\u0435\u0440\u0432\u0435\u0440\u044b\" \/>\n\t\t<meta property=\"og:type\" content=\"article\" \/>\n\t\t<meta property=\"og:title\" content=\"\ud83e\udd47\u0420\u0430\u0437\u0432\u0435\u0440\u0442\u044b\u0432\u0430\u043d\u0438\u0435 \u043f\u0440\u0438\u043b\u043e\u0436\u0435\u043d\u0438\u0439 \u043d\u0430 \u043d\u0435\u0441\u043a\u043e\u043b\u044c\u043a\u0438\u0445 \u043a\u043b\u0430\u0441\u0442\u0435\u0440\u0430\u0445 Kubernetes \u0441 Helm | ProHoster\" \/>\n\t\t<meta property=\"og:description\" content=\"\u041a\u0430\u043a Dailymotion \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u0435\u0442 Kubernetes: \u0440\u0430\u0437\u0432\u0435\u0440\u0442\u044b\u0432\u0430\u043d\u0438\u0435 \u043f\u0440\u0438\u043b\u043e\u0436\u0435\u043d\u0438\u0439 \u041c\u044b \u0432 Dailymotion \u043d\u0430\u0447\u0430\u043b\u0438 \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u0442\u044c Kubernetes \u0432.\" \/>\n\t\t<meta property=\"og:url\" content=\"https:\/\/prohoster.info\/hr\/blog\/administrirovanie\/razvertyvanie-prilozhenij-na-neskolkih-klasterah-kubernetes-s-helm\" \/>\n\t\t<meta property=\"og:image\" content=\"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg\" \/>\n\t\t<meta property=\"og:image:secure_url\" content=\"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg\" \/>\n\t\t<meta property=\"og:image:width\" content=\"350\" \/>\n\t\t<meta property=\"og:image:height\" content=\"350\" \/>\n\t\t<meta property=\"article:published_time\" content=\"2019-10-31T19:08:26+00:00\" \/>\n\t\t<meta property=\"article:modified_time\" content=\"2019-10-31T19:08:26+00:00\" \/>\n\t\t<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/prohoster\" \/>\n\t\t<meta property=\"article:author\" content=\"https:\/\/www.facebook.com\/prohoster\" \/>\n\t\t<!-- All in One SEO -->\n\n","aioseo_head_json":{"title":"\ud83e\udd47Implementacija aplikacija na nekoliko Kubernetes klastera s Helm | ProHoster","description":"Kako Dailymotion koristi Kubernetes: implementacija aplikacija. U Dailymotionu smo po\u010deli koristiti Kubernetes u.","canonical_url":"https:\/\/prohoster.info\/hr\/blog\/administrirovanie\/razvertyvanie-prilozhenij-na-neskolkih-klasterah-kubernetes-s-helm","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"hr_HR","og:site_name":"ProHoster | \u041a\u0443\u043f\u0438\u0442\u044c \u043d\u0430\u0434\u0435\u0436\u043d\u044b\u0439 \u0445\u043e\u0441\u0442\u0438\u043d\u0433 \u0434\u043b\u044f \u0441\u0430\u0439\u0442\u043e\u0432 \u0441 \u0437\u0430\u0449\u0438\u0442\u043e\u0439 \u043e\u0442 DDoS, VPS VDS \u0441\u0435\u0440\u0432\u0435\u0440\u044b","og:type":"article","og:title":"\ud83e\udd47\u0420\u0430\u0437\u0432\u0435\u0440\u0442\u044b\u0432\u0430\u043d\u0438\u0435 \u043f\u0440\u0438\u043b\u043e\u0436\u0435\u043d\u0438\u0439 \u043d\u0430 \u043d\u0435\u0441\u043a\u043e\u043b\u044c\u043a\u0438\u0445 \u043a\u043b\u0430\u0441\u0442\u0435\u0440\u0430\u0445 Kubernetes \u0441 Helm | ProHoster","og:description":"\u041a\u0430\u043a Dailymotion \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u0435\u0442 Kubernetes: \u0440\u0430\u0437\u0432\u0435\u0440\u0442\u044b\u0432\u0430\u043d\u0438\u0435 \u043f\u0440\u0438\u043b\u043e\u0436\u0435\u043d\u0438\u0439 \u041c\u044b \u0432 Dailymotion \u043d\u0430\u0447\u0430\u043b\u0438 \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u0442\u044c Kubernetes \u0432.","og:url":"https:\/\/prohoster.info\/hr\/blog\/administrirovanie\/razvertyvanie-prilozhenij-na-neskolkih-klasterah-kubernetes-s-helm","og:image":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:secure_url":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:width":350,"og:image:height":350,"article:published_time":"2019-10-31T19:08:26+00:00","article:modified_time":"2019-10-31T19:08:26+00:00","article:publisher":"https:\/\/www.facebook.com\/prohoster","article:author":"https:\/\/www.facebook.com\/prohoster"},"aioseo_meta_data":{"post_id":"35955","title":null,"description":null,"keywords":null,"keyphrases":null,"primary_term":null,"canonical_url":null,"og_title":null,"og_description":null,"og_object_type":"default","og_image_type":"default","og_image_url":null,"og_image_width":null,"og_image_height":null,"og_image_custom_url":null,"og_image_custom_fields":null,"og_video":null,"og_custom_url":null,"og_article_section":null,"og_article_tags":null,"twitter_use_og":false,"twitter_card":"default","twitter_image_type":"default","twitter_image_url":null,"twitter_image_custom_url":null,"twitter_image_custom_fields":null,"twitter_title":null,"twitter_description":null,"schema":{"blockGraphs":[],"customGraphs":[],"default":{"data":{"Article":[],"Course":[],"Dataset":[],"FAQPage":[],"Movie":[],"Person":[],"Product":[],"ProductReview":[],"Car":[],"Recipe":[],"Service":[],"SoftwareApplication":[],"WebPage":[]},"graphName":"","isEnabled":true},"graphs":[]},"schema_type":null,"schema_type_options":null,"pillar_content":false,"robots_default":true,"robots_noindex":false,"robots_noarchive":false,"robots_nosnippet":false,"robots_nofollow":false,"robots_noimageindex":false,"robots_noodp":false,"robots_notranslate":false,"robots_max_snippet":null,"robots_max_videopreview":null,"robots_max_imagepreview":"large","priority":null,"frequency":null,"local_seo":null,"seo_analyzer_scan_date":"2026-01-22 01:25:19","breadcrumb_settings":null,"limit_modified_date":false,"reviewed_by":null,"ai":null,"created":"2021-03-01 01:53:31","updated":"2026-01-22 01:25:19","focus_keyword":null,"additional_keywords":null,"truseo_locale":null},"gt_translate_keys":[{"key":"link","format":"url"}],"_links":{"self":[{"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/posts\/35955","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/comments?post=35955"}],"version-history":[{"count":0,"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/posts\/35955\/revisions"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/media?parent=35955"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/categories?post=35955"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/tags?post=35955"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}