{"id":34596,"date":"2019-10-31T21:59:16","date_gmt":"2019-10-31T18:59:16","guid":{"rendered":"https:\/\/prohoster.info\/blog\/zhivut-li-bazy-dannyh-v-kubernetes\/"},"modified":"2019-10-31T21:59:16","modified_gmt":"2019-10-31T18:59:16","slug":"zhivut-li-bazy-dannyh-v-kubernetes","status":"publish","type":"post","link":"https:\/\/prohoster.info\/ro\/blog\/administrirovanie\/zhivut-li-bazy-dannyh-v-kubernetes","title":{"rendered":"Exist\u0103 baze de date \u00een Kubernetes?","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

\"Exist\u0103
\n
\nIstoria a f\u0103cut ca industria IT s\u0103 se \u00eempart\u0103, \u00een mod tradi\u021bional, \u00een dou\u0103 tabere condi\u021bionate: cei \u201epentru\u201d \u0219i cei \u201e\u00eempotriv\u0103\u201d. Subiectul disputelor poate fi absolut aleator. Ce sistem de operare este mai bun: Windows sau Linux? S\u0103 ai un smartphone cu Android sau iOS? S\u0103 stochezi totul \u00een cloud sau s\u0103 folose\u0219ti unit\u0103\u021bi RAID cold storage \u0219i s\u0103 pui hard disk-urile \u00eentr-un seif? Au dreptul dezvoltatorii PHP s\u0103 fie numi\u021bi programatori? Aceste dispute sunt, uneori, de natur\u0103 exclusiv existen\u021bial\u0103 \u0219i nu au la baz\u0103 nimic altceva \u00een afar\u0103 de un interes pur sportiv.<\/p>\n

A\u0219a s-a \u00eent\u00e2mplat c\u0103, odat\u0103 cu apari\u021bia containerelor \u0219i a acestei dragi industrii cu Docker \u0219i, \u00een mod simbolic, Kubernetes, au \u00eenceput \u0219i disputele \u201epentru\u201d \u0219i \u201e\u00eempotriv\u0103\u201d cu privire la utilizarea noilor oportunit\u0103\u021bi \u00een diferite domenii ale backend-ului. (De la \u00eenceput, s\u0103 men\u021bion\u0103m c\u0103, de\u0219i cel mai frecvent \u00een aceast\u0103 argumenta\u021bie se va face referire la Kubernetes ca orchestrator, alegerea acestui instrument specific nu este crucial\u0103. \u00cen locul s\u0103u, po\u021bi alege orice alt\u0103 op\u021biune care \u021bi se pare mai convenabil\u0103 \u0219i familiar\u0103.)<\/p>\n

\u0218i, se p\u0103rea c\u0103 ar fi fost o simpl\u0103 disput\u0103 \u00eentre dou\u0103 p\u0103r\u021bi ale acelea\u0219i medalii. La fel de lipsit\u0103 de sens \u0219i implacabil\u0103 ca eterna confruntare Windows vs Linux, \u00een care oamenii ra\u021bionali exist\u0103, de fapt, undeva la mijloc. Totu\u0219i, \u00een cazul containeriz\u0103rii, lucrurile nu sunt at\u00e2t de simple. De obicei, \u00een astfel de dispute nu exist\u0103 o parte corect\u0103, dar \u00een cazul \u201ea folosi\u201d sau \u201ea nu folosi\u201d containere pentru stocarea bazelor de date, totul se \u00eentoarce pe dos. Pentru c\u0103, \u00een anumite sensuri, at\u00e2t sus\u021bin\u0103torii c\u00e2t \u0219i oponen\u021bii acestui tip de abordare au dreptate.
\n<\/a><\/noindex><\/p>\n

Latura Luminoas\u0103<\/h4>\n

\nO scurt\u0103 descriere a argumenta\u021biei Laturii Luminoase poate fi rezumat\u0103 \u00eentr-o fraz\u0103: \u201eSalut, 2k19 la u\u0219\u0103!\u201d Sun\u0103 ca un populism, f\u0103r\u0103 \u00eendoial\u0103, dar dac\u0103 ne aplec\u0103m asupra situa\u021biei \u00een detaliu, exist\u0103 anumite avantaje. Acestea le vom analiza acum.<\/p>\n

Presupunem c\u0103 ave\u021bi un proiect web mare. Acesta ar fi putut fi ini\u021bial construit pe baza unei abord\u0103ri de microservicii sau a ajuns la acest lucru printr-o evolu\u021bie treptat\u0103 \u2014 nu este foarte important, de fapt. A\u021bi distribuit proiectul pe microservicii separate, a\u021bi configurat orchestrarea, echilibrarea \u00eenc\u0103rc\u0103turii, scalarea. \u0218i acum, cu o con\u0219tiin\u021b\u0103 curat\u0103, savura\u021bi un mohito \u00een hamac \u00een timpul efectelor Habr \u00een loc s\u0103 ridica\u021bi serverele c\u0103zute. Dar trebuie s\u0103 fi\u021bi consecvent \u00een toate ac\u021biunile. Foarte adesea, doar aplica\u021bia \u00een sine \u2014 codul \u2014 este containerizat\u0103. \u0218i ce mai avem \u00een afar\u0103 de cod?<\/p>\n

Corect, datele. Inima oric\u0103rui proiect sunt datele sale: acestea pot fi o baz\u0103 de date tipic\u0103 \u2014 MySQL, Postgre, MongoDB, precum \u0219i stoc\u0103ri utilizate pentru c\u0103utare (ElasticSearch), stoc\u0103ri key-value pentru caching \u2014 de exemplu, redis, etc. Acum nu vom vorbi despre variantele gre\u0219ite de implementare a backend-ului, c\u00e2nd baza de date se pr\u0103bu\u0219e\u0219te din cauza interog\u0103rilor prost scrise, ci vom discuta despre asigurarea redundan\u021bei acestei baze de date sub \u00eenc\u0103rc\u0103tura clien\u021bilor. C\u00e2nd containeriz\u0103m aplica\u021bia noastr\u0103 \u0219i \u00eei permitem s\u0103 se scaleze liber pentru a gestiona orice num\u0103r de solicit\u0103ri, acest lucru cre\u0219te, \u00een mod inevitabil, \u0219i \u00eenc\u0103rc\u0103tura pe baza de date.<\/p>\n

De fapt, canalul de acces la baza de date \u0219i serverul pe care aceasta ruleaz\u0103 devin un g\u00e2t de \u00eendoire \u00een backendul nostru frumos containerizat. \u00cen acela\u0219i timp, principalul motiv pentru virtualizarea prin containere este mobilitatea \u0219i flexibilitatea structurii, care permit organizarea distribu\u021biei \u00eenc\u0103rc\u0103turii de v\u00e2rf \u00een \u00eentreaga infrastructur\u0103 disponibil\u0103 c\u00e2t mai eficient posibil. Asta \u00eenseamn\u0103 c\u0103, dac\u0103 nu containeriz\u0103m \u0219i nu distribui toate elementele existente ale sistemului pe un cluster, facem o gre\u0219eal\u0103 foarte serioas\u0103.<\/p>\n

Este mult mai logic s\u0103 clusteriz\u0103m nu doar aplica\u021bia, ci \u0219i serviciile care se ocup\u0103 cu stocarea datelor. Prin clusterizarea \u0219i desf\u0103\u0219urarea serverelor web, care opereaz\u0103 independent \u0219i distribuie sarcinile \u00eentre ele, \u00een k8s, abord\u0103m deja problema sincroniz\u0103rii datelor \u2014 de exemplu, comentariile la post\u0103ri, dac\u0103 lu\u0103m ca exemplu o platform\u0103 media sau un blog. \u00cen orice caz, se creeaz\u0103 o reprezentare a bazei de date, chiar \u0219i virtual\u0103, ca ExternalService \u00een cadrul cluster-ului. Problema este c\u0103 baza de date \u00eens\u0103 nu este \u00eenc\u0103 clusterizat\u0103 \u2014 serverele web desf\u0103\u0219urate \u00een kubernete iau informa\u021bii despre modific\u0103ri din baza noastr\u0103 static\u0103 de produc\u021bie, care ruleaz\u0103 separat.<\/p>\n

Sim\u021bi\u021bi un truc? Folosim k8s sau Swarm pentru a distribui sarcina \u0219i a evita c\u0103derea principalului server web<\/a>, dar nu facem asta pentru baza de date. Dar, dac\u0103 baza de date cade, nu are sens \u00een \u00eentreaga noastr\u0103 infrastructur\u0103 clusterizat\u0103 \u2014 ce folos avem de paginile web goale care returneaz\u0103 o eroare de acces la baza de date?<\/p>\n

De aceea trebuie clusterizate nu doar serverele web, a\u0219a cum se face \u00een mod obi\u0219nuit, ci \u0219i infrastructura bazei de date. Numai astfel putem asigura o structur\u0103 care func\u021bioneaz\u0103 complet \u00eentr-un singur sistem, dar care este, \u00een acela\u0219i timp, independent\u0103 una de cealalt\u0103. Chiar dac\u0103 jum\u0103tate din backend-ul nostru \u201ec\u00e2nd rupe\u201d sub sarcin\u0103 \u2014 cealalt\u0103 va supravie\u021bui, iar sistemul de sincronizare \u00eentre bazele de date din cadrul cluster-ului \u0219i capacitatea de scalare infiniti\u0103 \u0219i desf\u0103\u0219urarea de noi clustere ne va ajuta s\u0103 ajungem rapid la puterea necesar\u0103 \u2014 s\u0103 existe racks \u00een centrul de date.<\/p>\n

\u00cen plus, modelul distribuit \u00een clustere al bazei de date permite s\u0103 transporta\u021bi chiar \u0219i aceast\u0103 baz\u0103 de date acolo unde este necesar\u0103; dac\u0103 vorbim despre un serviciu global, este destul de ilogic s\u0103 rul\u0103m un cluster web undeva \u00een apropiere de San Francisco \u0219i s\u0103 transfer\u0103m pachete cer\u00e2nd baza de date din Podmoscovia \u0219i \u00eenapoi.<\/p>\n

De asemenea, containerizarea bazei de date permite construirea tuturor elementelor sistemului la acela\u0219i nivel de abstractizare. Ceea ce, la r\u00e2ndul s\u0103u, face posibil\u0103 gestionarea acestui sistem direct din cod, de c\u0103tre dezvoltatori, f\u0103r\u0103 a necesita implicarea activ\u0103 a administratorilor. Dac\u0103 dezvoltatorii consider\u0103 c\u0103 au nevoie de o anumit\u0103 baz\u0103 de date pentru un nou subproiect \u2014 este simplu! scriu un fi\u0219ier yaml, \u00eel \u00eencarc\u0103 \u00een cluster \u0219i gata.<\/p>\n

\u0218i, desigur, exploatarea intern\u0103 devine de zeci de ori mai u\u0219oar\u0103. Spune\u021bi-mi, c\u00e2teodat\u0103 a\u021bi \u00eenchis ochii c\u00e2nd un nou membru al echipei se b\u0103ga \u00een baza de date de produc\u021bie? Pe care, de fapt, o ave\u021bi doar pe una, care se afl\u0103 chiar acum \u00een utilizare? Desigur, cu to\u021bii suntem oameni maturi \u0219i avem, probabil, un backup recent, \u0219i mai departe \u2014 pe un raft cu mur\u0103turi ale bunicii \u0219i schiuri vechi \u2014 un alt backup, poate chiar \u00eentr-un depozit rece, pentru c\u0103 odat\u0103 oficiul dumneavoastr\u0103 a ars. Totu\u0219i, fiecare introducere a unui nou membru al echipei, care are acces la infrastructura de produc\u021bie \u0219i, desigur, la baza de date de produc\u021bie \u2014 este ca un cub de validol pentru to\u021bi cei din jur. Cine \u0219tie, poate acest novice este st\u00e2ngaci? Este \u00eenfrico\u0219\u0103tor, recunoa\u0219te\u021bi.<\/p>\n

Containerizarea \u0219i, prin urmare, topologia fizic\u0103 distribuit\u0103 a bazei de date a proiectului dumneavoastr\u0103 ajut\u0103 la evitarea acestor momente de validol. Nu ave\u021bi \u00eencredere \u00een novice? Bine! \u00cei vom ridica un cluster propriu pentru lucru \u0219i \u00eei vom deconecta de la celelalte clustere de baze de date \u2014 sincronizarea numai printr-un push manual \u0219i rota\u021bia sincron\u0103 a dou\u0103 chei (una pentru liderul echipei, cealalt\u0103 pentru admin). \u0218i toat\u0103 lumea este fericit\u0103.<\/p>\n

Acum a venit vremea s\u0103 ne transform\u0103m \u00een du\u0219mani ai clusteriz\u0103rii bazelor de date.<\/p>\n

Latura \u00eentunecat\u0103<\/h4>\n

\nReflect\u00e2nd la motivele pentru care nu ar trebui s\u0103 containeriz\u0103m baza de date \u0219i s\u0103 continu\u0103m s\u0103 o rul\u0103m pe un singur server central server<\/a>, s\u0103 nu ne cobor\u00e2m la retorica ortodoc\u0219ilor \u0219i afirma\u021biile de genul \u201estr\u0103mo\u0219ii au rulat bazele de date pe hardware \u0219i noi vom face la fel!\u201d \u00cen loc de aceasta, haide\u021bi s\u0103 \u00eencerc\u0103m s\u0103 imagin\u0103m o situa\u021bie \u00een care containerizarea ar aduce cu adev\u0103rat beneficii semnificative.<\/p>\n

S\u0103 fim de acord, proiectele care au cu adev\u0103rat nevoie de o baz\u0103 \u00een container se pot num\u0103ra pe degetele de la o m\u00e2n\u0103 a unui frezor nu foarte talentat. \u00cen majoritatea cazurilor, chiar \u0219i utilizarea k8s sau Docker Swarm este redundant\u0103 \u2014 destul de des, aceste instrumente sunt folosite pentru c\u0103 tehnologiile sunt la mod\u0103 \u0219i pentru a urma directivele \u201esupreme\u201d ale managerilor de a \u00eenc\u0103rca totul \u00een cloud \u0219i \u00een containere. Well, pentru c\u0103 acum este la mod\u0103 \u0219i toat\u0103 lumea face a\u0219a.<\/p>\n

\u00cen minimum jum\u0103tate din cazuri, utilizarea Kubernetes sau doar a Docker pe proiect este excesiv\u0103. Problema este c\u0103 nu toate echipele sau companiile de outsourcing angajate pentru a gestiona infrastructura clientului \u00ee\u0219i dau seama de acest lucru. Mai r\u0103u este atunci c\u00e2nd containerele sunt impuse, pentru c\u0103 asta cost\u0103 un anumit num\u0103r de monede clientului.<\/p>\n

\u00cen general, exist\u0103 o opinie conform c\u0103reia mafia Docker\/Kubernetes subjug\u0103 pur \u0219i simplu clien\u021bii care externalizeaz\u0103 aceste probleme de infrastructur\u0103. Deoarece, pentru a lucra cu clustere, sunt necesari ingineri care sunt capabili s\u0103 fac\u0103 acest lucru \u0219i care \u00een\u021beleg arhitectura solu\u021biei implementate. Am descris deja cazul nostru cu revista Republic \u2014 acolo am instruit echipa clientului s\u0103 lucreze \u00een realit\u0103\u021bile Kubernetes, iar to\u021bi au fost mul\u021bumi\u021bi. \u0218i a fost onorabil. De multe ori, \u201eimplementatorii\u201d k8s iau infrastructura clientului ostatice \u2014 deoarece acum doar ei \u00een\u021beleg cum funcioneaz\u0103 totul, iar de partea clientului nu exist\u0103 speciali\u0219ti.<\/p>\n

\u0218i acum imagina\u021bi-v\u0103 c\u0103 astfel ced\u0103m nu doar partea de server web outsourcing-ului, ci \u0219i \u00eentre\u021binerea bazei de date. Am spus c\u0103 baza de date este inima, iar pierderea inimii este fatal\u0103 pentru orice organism viu. Pe scurt, perspectivele nu sunt cele mai bune. A\u0219a c\u0103, \u00een loc de hype-ul Kubernetes, multe proiecte ar trebui pur \u0219i simplu s\u0103 nu fie zg\u00e2rcite la un tarif normal pe AWS, care ar rezolva toate problemele cu \u00eenc\u0103rcarea pe site-ul\/proiectul lor. Dar AWS nu mai este la mod\u0103, iar imaginea este mai scump\u0103 dec\u00e2t banii \u2014 din p\u0103cate, \u0219i \u00een mediul IT.<\/p>\n

Ok. Poate c\u0103 clasterizarea este \u00eentr-adev\u0103r necesar\u0103 pentru proiect, dar dac\u0103 cu aplica\u021biile stateless totul este clar, cum ar trebui s\u0103 organiz\u0103m o conectivitate adecvat\u0103 pentru o baz\u0103 de date clasterizat\u0103?<\/p>\n

C\u00e2nd vorbim despre o solu\u021bie ingineresc\u0103 f\u0103r\u0103 \u00eentreruperi, care este considerat\u0103 a fi tranzi\u021bia la k8s, cea mai mare durere de cap este replicarea datelor \u00eentr-o baz\u0103 de date clusterizat\u0103. Unele SGBD-uri sunt destul de tolerante fa\u021b\u0103 de distribuirea datelor \u00eentre instan\u021bele lor. Multe altele, \u00eens\u0103, nu sunt at\u00e2t de primitoare. De obicei, principalul argument \u00een alegerea SGBD-ului pentru proiectul nostru nu este neap\u0103rat capacitatea de a se replica cu resurse \u0219i costuri de inginerie minime. \u00cen special dac\u0103 proiectul nu a fost ini\u021bial planificat ca fiind microserviciu, ci a evoluat \u00een aceast\u0103 direc\u021bie. <\/p>\n

Credem c\u0103 nu trebuie s\u0103 explic\u0103m despre viteza discurilor de re\u021bea \u2014 sunt lente. Adic\u0103, \u00een cazul \u00een care este necesar s\u0103 repornim o instan\u021b\u0103 de SGBD undeva unde sunt, de exemplu, mai multe puteri de procesare sau memorie RAM disponibil\u0103, totu\u0219i nu avem aceast\u0103 posibilitate. Ne vom lovi foarte repede de limita de performan\u021b\u0103 a subsistemului de discuri virtualizate. Prin urmare, SGBD-ul trebuie s\u0103 fie legat de propriul s\u0103u set personal de ma\u0219ini, aflate \u00een imediata apropiere. Alternativ, va trebui s\u0103 g\u0103sim o sincronizare rapid\u0103 a datelor c\u0103tre rezervele preconizate. <\/p>\n

Continu\u00e2nd tema sistemelor de fi\u0219iere virtuale: Volumele Docker, din p\u0103cate, nu sunt lipsite de probleme. \u00cen general, \u00een ceea ce prive\u0219te stocarea pe termen lung \u0219i fiabil\u0103 a datelor, ne-ar pl\u0103cea s\u0103 folosim scheme tehnice c\u00e2t mai simple. Ad\u0103ugarea unei noi straturi de abstractizare din sistemul de fi\u0219iere al containerului \u00een sistemul de fi\u0219iere al gazdei p\u0103rinte reprezint\u0103 deja un risc \u00een sine. Dar atunci c\u00e2nd apar dificult\u0103\u021bi cu translatrea datelor \u00eentre aceste straturi \u00een timpul func\u021bion\u0103rii sistemului de contentie a containerelor, devine chiar mai problematic. \u00cen prezent, majoritatea problemelor bine cunoscute de umanitate pare c\u0103 sunt eradicate. Dar, dup\u0103 cum \u00een\u021belege\u021bi, cu c\u00e2t mecanismul este mai complex, cu at\u00e2t este mai predispus s\u0103 se r\u0103t\u0103ceasc\u0103.<\/p>\n

\u00cen lumina tuturor acestor \u201eaventuri\u201d, este mult mai rentabil \u0219i mai simplu s\u0103 p\u0103strezi baza de date \u00eentr-un singur loc, iar chiar dac\u0103 ai nevoie de containerizarea aplica\u021biei \u2014 las-o s\u0103 ruleze de una singur\u0103, ob\u021bin\u00e2nd conexiunea simultan\u0103 cu baza de date printr-o poart\u0103 de distribu\u021bie, care va fi citit\u0103 \u0219i scris\u0103 doar o dat\u0103 \u0219i \u00eentr-un singur loc. Aceast\u0103 abordare reduce la minimum \u0219ansele de erori \u0219i desincroniz\u0103ri.<\/p>\n

Unde vrem s\u0103 ajungem? C\u0103 containerizarea bazei de date este adecvat\u0103 acolo unde exist\u0103 o necesitate real\u0103. Nu po\u021bi s\u0103 \u00eempingi o baz\u0103 de date de tip full-app \u0219i s\u0103 o rulezi de parc\u0103 ai avea dou\u0103 zeci de microservicii \u2014 a\u0219a nu func\u021bioneaz\u0103. Trebuie s\u0103 \u00een\u021belegem foarte clar acest lucru.<\/p>\n

\u00cen loc de concluzie<\/h4>\n

\nDac\u0103 a\u0219tep\u021bi un r\u0103spuns clar de tipul \u201es\u0103 virtualiz\u0103m sau nu baza de date\u201d, ne pare r\u0103u: nu va exista. Pentru c\u0103, \u00een crearea oric\u0103rei solu\u021bii infrastructurale, trebuie s\u0103 te ghidezi nu dup\u0103 mod\u0103 \u0219i progres, ci \u00een primul r\u00e2nd, dup\u0103 bunul sim\u021b.<\/p>\n

Exist\u0103 proiecte pentru care principiile \u0219i instrumentele care vin odat\u0103 cu Kubernetes se potrivesc perfect, iar \u00een astfel de proiecte, pacea se stabile\u0219te cel pu\u021bin \u00een domeniul backend. \u0218i exist\u0103 proiecte care nu necesit\u0103 containerizare, ci o infrastructur\u0103 de server normal\u0103, deoarece pur \u0219i simplu nu pot s\u0103 se scalzeze sub modelul de cluster microservicii, altfel vor e\u0219ua.
\n
Sursa: habr.com<\/a><\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"

\u041a\u0430\u043a-\u0442\u043e \u0442\u0430\u043a \u0438\u0441\u0442\u043e\u0440\u0438\u0447\u0435\u0441\u043a\u0438 \u0441\u043b\u043e\u0436\u0438\u043b\u043e\u0441\u044c, \u0447\u0442\u043e IT-\u0438\u043d\u0434\u0443\u0441\u0442\u0440\u0438\u044f \u043f\u043e \u043b\u044e\u0431\u043e\u043c\u0443 \u043f\u043e\u0432\u043e\u0434\u0443 \u0440\u0430\u0437\u0431\u0438\u0432\u0430\u0435\u0442\u0441\u044f \u043d\u0430 \u0434\u0432\u0430 \u0443\u0441\u043b\u043e\u0432\u043d\u044b\u0445 \u043b\u0430\u0433\u0435\u0440\u044f: \u043a\u043e\u0442\u043e\u0440\u044b\u0435 \u00ab\u0437\u0430\u00bb \u0438 \u043a\u043e\u0442\u043e\u0440\u044b\u0435 \u00ab\u043f\u0440\u043e\u0442\u0438\u0432\u00bb. \u041f\u0440\u0438\u0447\u0435\u043c \u043f\u0440\u0435\u0434\u043c\u0435\u0442 \u0441\u043f\u043e\u0440\u043e\u0432 \u043c\u043e\u0436\u0435\u0442 \u0431\u044b\u0442\u044c \u0430\u0431\u0441\u043e\u043b\u044e\u0442\u043d\u043e \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u043b\u044c\u043d\u044b\u043c. \u041a\u0430\u043a\u0430\u044f \u041e\u0421 \u043b\u0443\u0447\u0448\u0435: Win \u0438\u043b\u0438 Linux? \u041d\u0430 \u0441\u043c\u0430\u0440\u0442\u0444\u043e\u043d\u0435 Android \u0438\u043b\u0438 iOS? \u0425\u0440\u0430\u043d\u0438\u0442\u044c \u0432\u0441\u0435 \u0432 \u043e\u0431\u043b\u0430\u043a\u0430\u0445 \u0438\u043b\u0438 \u0437\u0430\u043b\u0438\u0432\u0430\u0442\u044c \u043d\u0430 \u0445\u043e\u043b\u043e\u0434\u043d\u044b\u0435 RAID-\u0445\u0440\u0430\u043d\u0438\u043b\u0438\u0449\u0430 \u0438 \u043a\u043b\u0430\u0441\u0442\u044c \u0432\u0438\u043d\u0442\u044b \u0432 \u0441\u0435\u0439\u0444? \u0418\u043c\u0435\u044e\u0442 \u043b\u0438 PHP-\u0448\u043d\u0438\u043a\u0438 \u043f\u0440\u0430\u0432\u043e […]<\/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-34596","post","type-post","status-publish","format-standard","hentry","category-administrirovanie"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\n","aioseo_head_json":{"title":"\ud83e\udd47Tr\u0103iesc baze de date \u00een Kubernetes? | ProHoster","description":"Astfel, istoric s-a realizat c\u0103 industria IT.","canonical_url":"https:\/\/prohoster.info\/ro\/blog\/administrirovanie\/zhivut-li-bazy-dannyh-v-kubernetes","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"ro_RO","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\u0416\u0438\u0432\u0443\u0442 \u043b\u0438 \u0431\u0430\u0437\u044b \u0434\u0430\u043d\u043d\u044b\u0445 \u0432 Kubernetes? | ProHoster","og:description":"\u041a\u0430\u043a-\u0442\u043e \u0442\u0430\u043a \u0438\u0441\u0442\u043e\u0440\u0438\u0447\u0435\u0441\u043a\u0438 \u0441\u043b\u043e\u0436\u0438\u043b\u043e\u0441\u044c, \u0447\u0442\u043e IT-\u0438\u043d\u0434\u0443\u0441\u0442\u0440\u0438\u044f \u043f\u043e.","og:url":"https:\/\/prohoster.info\/ro\/blog\/administrirovanie\/zhivut-li-bazy-dannyh-v-kubernetes","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":"2019-10-31T18:59:16+00:00","article:modified_time":"2019-10-31T18:59:16+00:00","article:publisher":"https:\/\/www.facebook.com\/prohoster","article:author":"https:\/\/www.facebook.com\/prohoster"},"aioseo_meta_data":{"post_id":"34596","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":"2026-02-09 21:46:19","breadcrumb_settings":null,"limit_modified_date":false,"reviewed_by":null,"ai":null,"created":"2021-03-01 02:18:36","updated":"2026-02-09 21:46:19","focus_keyword":null,"additional_keywords":null,"truseo_locale":null},"gt_translate_keys":[{"key":"link","format":"url"}],"_links":{"self":[{"href":"https:\/\/prohoster.info\/ro\/wp-json\/wp\/v2\/posts\/34596","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/ro\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/ro\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/ro\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/ro\/wp-json\/wp\/v2\/comments?post=34596"}],"version-history":[{"count":2,"href":"https:\/\/prohoster.info\/ro\/wp-json\/wp\/v2\/posts\/34596\/revisions"}],"predecessor-version":[{"id":159764,"href":"https:\/\/prohoster.info\/ro\/wp-json\/wp\/v2\/posts\/34596\/revisions\/159764"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/ro\/wp-json\/wp\/v2\/media?parent=34596"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/ro\/wp-json\/wp\/v2\/categories?post=34596"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/ro\/wp-json\/wp\/v2\/tags?post=34596"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}