{"id":33773,"date":"2019-10-31T21:54:36","date_gmt":"2019-10-31T18:54:36","guid":{"rendered":"https:\/\/prohoster.info\/blog\/rezervirovanie-v-kubernetes-ono-sushhestvuet\/"},"modified":"2019-10-31T21:54:36","modified_gmt":"2019-10-31T18:54:36","slug":"rezervirovanie-v-kubernetes-ono-sushhestvuet","status":"publish","type":"post","link":"https:\/\/prohoster.info\/sq\/blog\/administrirovanie\/rezervirovanie-v-kubernetes-ono-sushhestvuet","title":{"rendered":"Rezervimi n\u00eb Kubernetes: ekziston","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

M\u00eb quajn\u00eb Sergey, jam nga kompania ITSumma, dhe dua t'ju tregoj si ne e qasemi rezervimit n\u00eb Kubernetes. Koh\u00ebt e fundit kam punuar shum\u00eb n\u00eb k\u00ebshillimin p\u00ebr zbatimin e zgjidhjeve t\u00eb ndryshme devops p\u00ebr ekipe t\u00eb ndryshme, dhe, ve\u00e7an\u00ebrisht, kam punuar ngusht\u00ebsisht n\u00eb projekte q\u00eb p\u00ebrdorin K8s. N\u00eb konferenc\u00ebn Uptime day 4, e cila ishte e fokusuar n\u00eb rezervimin n\u00eb arkitektura komplekse, kam mbajtur nj\u00eb ligj\u00ebrat\u00eb mbi rezervimin e \"kubik\u00ebs\", dhe ja nj\u00eb p\u00ebrmbledhje e lir\u00eb e saj. Por m\u00eb par\u00eb d\u00ebshiroj t\u00eb paralajm\u00ebroj se ky nuk \u00ebsht\u00eb nj\u00eb udh\u00ebzues i drejtp\u00ebrdrejt\u00eb p\u00ebr veprim, por m\u00eb tep\u00ebr nj\u00eb p\u00ebrmbledhje e mendimeve mbi tem\u00ebn e p\u00ebrmendur.<\/p>\n

\"Rezervimi
\n
\nN\u00eb parim, monitorimi dhe rezervimi jan\u00eb dy mjetet kryesore p\u00ebr rritjen e q\u00ebndrueshm\u00ebris\u00eb s\u00eb \u00e7do projekti. Por, ndoshta do t\u00eb thoni, n\u00eb kuber gjith\u00e7ka balancohet vet\u00eb, gjith\u00e7ka shkall\u00ebzohet vet\u00eb, dhe n\u00ebse ndodh di\u00e7ka - ajo ngrihet vet\u00eb... K\u00ebshtu q\u00eb, gjat\u00eb nj\u00eb hulumtimi t\u00eb shkurt\u00ebr t\u00eb tem\u00ebs, p\u00ebrgjigjja n\u00eb internet p\u00ebr pyetjen se si qasemi n\u00eb rezervimin K8s ishte \"pse?\" Shumica mendojn\u00eb se kuber \u00ebsht\u00eb nj\u00eb gj\u00eb magjike q\u00eb heq \u00e7do problem infrastruktural dhe b\u00ebn q\u00eb projekti t\u00eb mos bjer\u00eb kurr\u00eb. Por... bota nuk \u00ebsht\u00eb ashtu si\u00e7 duket.
\n<\/a><\/noindex>
\nSi e qas\u00ebm ne procesin e rezervimit m\u00eb par\u00eb? Kishim platforma identike p\u00ebr akomodim - ose ishin virtuale, ose ishin t\u00eb metalit server\u00ebt<\/a>, t\u00eb cilave u aplikonim tri praktika baz\u00eb: <\/p>\n

    \n
  1. sinkronizimi i kodit dhe statik\u00ebs<\/li>\n
  2. sinkronizimi i konfigurimeve<\/li>\n
  3. replikimi i bazave t\u00eb dh\u00ebnash<\/li>\n<\/ol>\n

    \nDhe vuala: n\u00eb \u00e7do moment ne kalojm\u00eb n\u00eb platform\u00ebn rezerv\u00eb, t\u00eb gjith\u00eb jan\u00eb t\u00eb lumtur, ne q\u00ebndrojm\u00eb dhe shp\u00ebrndaheni. <\/p>\n

    \"Rezervimi
    \n
    \n\u00c7far\u00eb na ofrojn\u00eb p\u00ebr t\u00eb rritur disponueshm\u00ebrin\u00eb e vazhdueshme t\u00eb aplikacionit ton\u00eb kubernetes? E para, p\u00ebr t\u00eb cil\u00ebn flet dokumentacioni jozyrtar \u2014 \u00ebsht\u00eb vendosja e shum\u00eb makinerive, t\u00eb b\u00ebsh shum\u00eb mastera \u2014 numri i tyre duhet t\u00eb plot\u00ebsoj\u00eb kushtet p\u00ebr t\u00eb arritur kuorum brenda klasterit, dhe q\u00eb n\u00eb \u00e7do master t\u00eb jet\u00eb ngritur etcd, api, MC, scheduler\u2026 Dhe, duket sikur gjith\u00e7ka \u00ebsht\u00eb e mrekullueshme: n\u00eb rast se disa node t\u00eb pun\u00ebs ose masterat shkojn\u00eb jasht\u00eb funksionit, klasteri yn\u00eb p\u00ebrshtatet, dhe aplikacioni vazhdon t\u00eb punoj\u00eb. P\u00ebrs\u00ebri duket si magji! Por shpesh klasteri yn\u00eb ndodhet brenda nj\u00eb qendre t\u00eb vetme t\u00eb p\u00ebrpunimit t\u00eb t\u00eb dh\u00ebnave dhe kjo mund t\u00eb shkaktoj\u00eb disa pyetje. \u00c7far\u00eb ndodh n\u00ebse nj\u00eb ekskavator mb\u00ebrrin dhe g\u00ebrmon kabllin, godet nj\u00eb rrufe, ndodh nj\u00eb p\u00ebrmbytje universale? Gjith\u00e7ka \u00ebsht\u00eb shkat\u00ebrruar, klasteri yn\u00eb nuk ekziston m\u00eb. Si t\u00eb qasemi ndaj rezervimit duke marr\u00eb parasysh k\u00ebt\u00eb aspekt t\u00eb problemit? <\/p>\n

    S\u00eb pari, ju duhet t\u00eb keni nj\u00eb klaster tjet\u00ebr n\u00eb rezerv\u00eb aktive, pra nj\u00eb klaster, n\u00eb t\u00eb cilin mund t\u00eb kaloni n\u00eb \u00e7do moment. N\u00eb t\u00eb nj\u00ebjt\u00ebn koh\u00eb, nga pik\u00ebpamja e kubernetes, infrastruktura duhet t\u00eb jet\u00eb plot\u00ebsisht identike. K\u00ebshtu q\u00eb n\u00ebse ka ndonj\u00eb plugin jostandarde p\u00ebr pun\u00ebn me sistemin e skedar\u00ebve, zgjidhje t\u00eb personalizuara p\u00ebr ingress, ato duhet t\u00eb jen\u00eb plot\u00ebsisht identike n\u00eb dy (apo tre, ose dhjet\u00eb, k\u00ebtu deri n\u00eb sa para dhe forc\u00eb e administrat\u00ebs) klaster\u00ebt tuaj. \u00cbsht\u00eb e nevojshme t\u00eb p\u00ebrcaktoni qart\u00eb dy grupe aplikacionesh (implementimesh, statefulset, daemonset, cronjob, etj): cilat prej tyre mund t\u00eb punojn\u00eb vazhdimisht n\u00eb rezerv\u00eb dhe cilat \u00ebsht\u00eb m\u00eb mir\u00eb t\u00eb mos i aktivizoni deri n\u00eb kalimin e drejtp\u00ebrdrejt\u00eb. <\/p>\n

    Pra, a duhet q\u00eb klasteri yn\u00eb i rezerv\u00ebs t\u00eb jet\u00eb plot\u00ebsisht identik me klasterin ton\u00eb aktiv? Jo. N\u00ebse m\u00eb par\u00eb, n\u00eb kuad\u00ebr t\u00eb pun\u00ebs me projekte monolitike dhe infrastruktur\u00eb harduerike, mbanim nj\u00eb ambient praktikisht identik, n\u00eb kuad\u00ebr t\u00eb kubernetes, un\u00eb besoj se kjo nuk duhet t\u00eb ndodh\u00eb. Le t\u00eb shqyrtojm\u00eb se pse.<\/p>\n

    P\u00ebr shembull, le t\u00eb fillojm\u00eb me entitetet baz\u00eb t\u00eb Kubernetes \u2014 deployments \u2014 ato duhet t\u00eb jen\u00eb identike. Aplikimet duhet t\u00eb jen\u00eb t\u00eb ekzekutuara n\u00eb m\u00ebnyr\u00eb q\u00eb n\u00eb \u00e7do moment t\u00eb mund t\u00eb kapin p\u00ebrpunimin e trafikut dhe t\u00eb lejojn\u00eb q\u00eb projekti yn\u00eb t\u00eb vazhdoj\u00eb t\u00eb ekzistoj\u00eb. Kur flasim p\u00ebr skedar\u00ebt e konfigurimit, duhet t\u00eb shohim n\u00ebse ata duhet t\u00eb jen\u00eb identike apo jo. Pra, n\u00ebse ne, njer\u00ebzit e men\u00e7ur, nuk p\u00ebrdorim asnj\u00eb substanc\u00eb t\u00eb ndaluar dhe nuk ruajm\u00eb baz\u00ebn n\u00eb K8s, at\u00ebher\u00eb n\u00eb configmaps duhet t\u00eb kemi cil\u00ebsimet e qasjeve n\u00eb baz\u00ebn operative (pro\u00e7esi i rezervimit t\u00eb s\u00eb cil\u00ebs \u00ebsht\u00eb formuar ve\u00e7mas). Ndaj p\u00ebr t\u00eb siguruar qasjet n\u00eb kopjen rezerv\u00eb t\u00eb baz\u00ebs s\u00eb t\u00eb dh\u00ebnave, ne duhet t\u00eb kemi nj\u00eb sked\u00eb konfigurimi t\u00eb ve\u00e7ant\u00eb (configmap). Po ashtu, punojm\u00eb me secret\u2019\u00ebt: fjal\u00ebkalimet p\u00ebr qasje n\u00eb baz\u00eb, \u00e7el\u00ebsat e API; n\u00eb \u00e7do moment mund t\u00eb funksionoj\u00eb ose secret operativ, ose rezerv\u00eb. Pra, tashm\u00eb kemi dy entitete Kubernetes, versionet rezerv\u00eb t\u00eb t\u00eb cilave nuk duhet t\u00eb jen\u00eb identike me ato operative. Entiteti tjet\u00ebr mbi t\u00eb cilin duhet t\u00eb ndalemi \u00ebsht\u00eb cronjob. Cronjob\u2019\u00ebt n\u00eb rezerv\u00eb nuk duhet asnj\u00ebher\u00eb t\u00eb jen\u00eb identike me grupin e cronjob\u2019\u00ebve t\u00eb klasterit prodhues! N\u00ebse ngrem\u00eb nj\u00eb klaster rezerv\u00eb dhe e ngrem\u00eb plot\u00ebsisht me t\u00eb gjith\u00eb cronjob\u2019\u00ebt e aktivizuar, p\u00ebr shembull, njer\u00ebzit do t\u00eb marrin dy mesazhe nj\u00ebkoh\u00ebsisht n\u00eb vend t\u00eb nj\u00ebi. Ose ndonj\u00eb sinkronizim t\u00eb dh\u00ebnash me burime t\u00eb jashtme do t\u00eb ndodh\u00eb dy her\u00eb, duke rezultuar k\u00ebshtu n\u00eb shqet\u00ebsime dhe ankthe. <\/p>\n

    \"Rezervimi
    \n
    \nPor si na propozojn\u00eb t\u00eb organizojm\u00eb klasterin rezerv\u00eb njer\u00ebzit nga interneti? Nj\u00eb p\u00ebrgjigje e dyt\u00eb m\u00eb e njohur pas \"pse?\" \u2014 p\u00ebrdorimi i Federat\u00ebs Kubernetes. <\/p>\n

    \u00c7far\u00eb \u00ebsht\u00eb kjo? \u00cbsht\u00eb, le t\u00eb themi, nj\u00eb metakluster i madh. N\u00ebse e imagjinojm\u00eb arkitektur\u00ebn e kub\u00ebs \u2014 ku kemi nj\u00eb master, disa node \u2014 at\u00ebher\u00eb, nga pik\u00ebpamja e federat\u00ebs, gjithashtu kemi nj\u00eb master dhe disa node, vet\u00ebm se \u00e7do node \u00ebsht\u00eb nj\u00eb kluster i ve\u00e7ant\u00eb. Pra, punojm\u00eb me entitetet e nj\u00ebjta, me primitivat e nj\u00ebjt\u00eb, si me nj\u00eb kub\u00eb t\u00eb vetme, vet\u00ebm se manipulojm\u00eb jo me makinat tona fizike, por me t\u00eb gjith\u00eb klaster\u00ebt. N\u00eb kuad\u00ebr t\u00eb federat\u00ebs, kemi nj\u00eb sinkronizim t\u00eb plot\u00eb t\u00eb resurset federative nga prind\u00ebrit te pasardh\u00ebsit. P\u00ebr shembull, n\u00ebse kemi nisur nj\u00eb deploy n\u00ebp\u00ebrmjet federat\u00ebs \u2014 ai do t\u00eb implementohet n\u00eb \u00e7do kluster dyt\u00ebsor. N\u00ebse marrim ndonj\u00eb configmap, sekret, e shp\u00ebrndajm\u00eb at\u00eb n\u00ebp\u00ebrmjet federat\u00ebs \u2014 ai do t\u00eb shp\u00ebrndahet n\u00eb t\u00eb gjith\u00eb klaster\u00ebt tan\u00eb dyt\u00ebsor\u00eb; n\u00eb t\u00eb nj\u00ebjt\u00ebn koh\u00eb, federata lejon personalizimin e resurseve tona te f\u00ebmij\u00ebt. Pra, mor\u00ebm nj\u00eb configmap, e implementuam at\u00eb n\u00ebp\u00ebrmjet federat\u00ebs dhe pastaj, n\u00ebse na nevojitet t\u00eb b\u00ebjm\u00eb ndonj\u00eb rregullim n\u00eb klaster\u00ebt specifik\u00eb, shkojm\u00eb dhe e rregullojm\u00eb n\u00eb nj\u00eb kluster t\u00eb ve\u00e7ant\u00eb, dhe ky ndryshim nuk do t\u00eb sinkronizohet askund. <\/p>\n

    Kubernetes Federation \u2014 \u00ebsht\u00eb nj\u00eb mjet q\u00eb ekziston prej jo shum\u00eb cot\u00eb dhe nuk mb\u00ebshtet t\u00eb gjith\u00eb fondin e burimeve q\u00eb ofron vet\u00eb K8s: n\u00eb momentin e publikimit t\u00eb nj\u00ebr\u00ebs nga versionet e para t\u00eb dokumentacionit u tha se mb\u00ebshteteshin vet\u00ebm config-maps, deployment p\u00ebr replicat-set, ingress. Sekretet nuk mb\u00ebshteteshin dhe puna me volume po ashtu nuk mb\u00ebshteteshin. Nj\u00eb fond i tep\u00ebr i kufizuar. Sidomos n\u00ebse ne duam t\u00eb arg\u00ebtohemi, \u2014 p\u00ebr shembull, n\u00ebp\u00ebrmjet custom resource definition p\u00ebr t\u00eb kaluar burime tona n\u00eb Kubernetes, \u2014 n\u00eb federat\u00eb nuk do t\u2019i arrijm\u00eb dot. Pra, si t\u00eb thuash\u2026 zgjidhje q\u00eb duket shum\u00eb e v\u00ebrtet\u00eb, por na b\u00ebn t\u00eb gjejm\u00eb raste p\u00ebr ta gjuajtur veten n\u00eb k\u00ebmb\u00eb her\u00eb pas here. N\u00eb an\u00ebn tjet\u00ebr, federata lejon t\u00eb menaxhojm\u00eb flexibilisht replicat-in ton\u00eb. P\u00ebr shembull, ne duam q\u00eb t\u00eb kemi 10 replika t\u00eb aplikacionit ton\u00eb, dhe si parazgjedhje federata do ta ndaj\u00eb k\u00ebt\u00eb num\u00ebr proporcionalisht mes numrit t\u00eb klastereve. Dhe kjo gjithashtu mund t\u00eb konfigurohet! Pra, mund t\u00eb p\u00ebrcaktojm\u00eb q\u00eb n\u00eb klasterin e prodhimit duhen mbajtur 6 replika t\u00eb aplikacionit ton\u00eb, dhe n\u00eb klasterin rezerv\u00eb, p\u00ebr t\u00eb kursyer burimet, ose p\u00ebr ndonj\u00eb arg\u00ebtim personal \u2014 vet\u00ebm 4 replika t\u00eb aplikacionit ton\u00eb. Kjo \u00ebsht\u00eb gjithashtu mjaft e p\u00ebrshtatshme. Por me federat\u00ebn na duhet t\u00eb p\u00ebrdorim disa zgjidhje t\u00eb reja, t\u00eb instalojm\u00eb di\u00e7ka n\u00eb l\u00ebvizje, t\u00eb b\u00ebjm\u00eb pak m\u00eb shum\u00eb mendime\u2026 <\/p>\n

    A mund t\u00eb afrohemi n\u00eb procesin e rezervimit t\u00eb Kubernetesit ndonj\u00ebher\u00eb m\u00eb thjesht? \u00c7far\u00eb mjete kemi n\u00eb dispozicion p\u00ebr ne?<\/p>\n

    S\u00eb pari, gjithmon\u00eb kemi nj\u00eb sistem ci\/cd, q\u00eb do t\u00eb thot\u00eb se ne nuk shkojm\u00eb dorazi dhe nuk shkruajm\u00eb n\u00eb server\u00eb create\/apply. Sistemi gjeneron yaml\u2019\u00ebt p\u00ebr kontejner\u00ebt tan\u00eb. <\/p>\n

    S\u00eb dyti, ekzistojn\u00eb disa klaster\u00eb, kemi nj\u00eb ose disa (n\u00ebse jemi t\u00eb zgjuar) registries, q\u00eb gjithashtu i kemi rezervuar. Dhe ka nj\u00eb utilitar t\u00eb jasht\u00ebzakonsh\u00ebm, kubectl, i cili mund t\u00eb punoj\u00eb me disa klaster\u00eb nj\u00ebkoh\u00ebsisht. <\/p>\n

    \"Rezervimi
    \n
    \nPra ndiheni: sipas mendimit tim, zgjidhja m\u00eb e thjesht\u00eb dhe e sakt\u00eb p\u00ebr nd\u00ebrtimin e nj\u00eb klasteri rezerv\u00eb \u00ebsht\u00eb nj\u00eb implementim paralel primitiv. Ka nj\u00eb pip\u00ebline n\u00eb sistemin ci\/cd; s\u00eb pari nd\u00ebrlidhim kontejner\u00ebt tan\u00eb, i testojm\u00eb dhe i vendosim aplikacionet p\u00ebrmes kubectl n\u00eb disa klaster\u00eb t\u00eb pavarur. Mund t\u00eb nd\u00ebrtojm\u00eb shp\u00ebrndarje paralelisht n\u00eb disa klaster\u00eb. N\u00eb p\u00ebrputhje, ne gjithashtu e zgjidhim d\u00ebrgimin e konfigurimeve n\u00eb k\u00ebt\u00eb faz\u00eb. Mund t\u00eb p\u00ebrcaktojm\u00eb paraprakisht nj\u00eb grup konfigurimesh p\u00ebr klasterin ton\u00eb n\u00eb prodhim, nj\u00eb grup konfigurimesh p\u00ebr klasterin rezerv\u00eb dhe n\u00eb nivelin e sistemit ci\/cd t\u00eb shp\u00ebrndajm\u00eb mjedisin e prodhimit n\u00eb klasterin e prodhimit, mjedisin rezerv\u00eb - n\u00eb klasterin rezerv\u00eb. N\u00eb krahasim me federat\u00ebn, nuk na duhet t\u00eb shkojm\u00eb pas p\u00ebrcaktimit t\u00eb burimit federal n\u00eb \u00e7do klaster t\u00eb bij\u00eb dhe t\u00eb rinegocjojm\u00eb ndonj\u00eb gj\u00eb. Ne e kemi b\u00ebr\u00eb k\u00ebt\u00eb paraprakisht. Sa mir\u00eb jemi.<\/p>\n

    Por\u2026 ka\u2026 si\u00e7 kam shkruar, ka dy \u2018rr\u00ebnj\u00eb t\u00eb t\u00eb gjitha t\u00eb k\u00ebqijave\u2019. E para \u00ebsht\u00eb sistemi i skedar\u00ebve. Ka ndonj\u00eb PV, ose ne p\u00ebrdorim nj\u00eb ruajtje t\u00eb jashtme. N\u00ebse ne ruajm\u00eb skedar\u00ebt brenda klasterit, at\u00ebher\u00eb duhet t\u00eb veprojm\u00eb sipas praktikave t\u00eb vjetra, q\u00eb kan\u00eb mbetur nga koha e infrastrukturave fizike: p\u00ebr shembull, t\u00eb sinkronizojm\u00eb me lsync. Ose ndonj\u00eb bast i preferuar nga ju. Zgjidhim gjith\u00e7ka n\u00eb makinat e tjera dhe vazhdojm\u00eb t\u00eb jetojm\u00eb.<\/p>\n

    E dyta, dhe, n\u00eb t\u00eb v\u00ebrtet\u00eb, nj\u00eb pik\u00eb m\u00eb e r\u00ebnd\u00ebsishme pengese - baza e t\u00eb dh\u00ebnave. N\u00ebse jemi njer\u00ebz t\u00eb men\u00e7ur dhe nuk e mbajm\u00eb baz\u00ebn e t\u00eb dh\u00ebnave n\u00eb kuber, at\u00ebher\u00eb procesi i rezervimit t\u00eb t\u00eb dh\u00ebnave sipas t\u00eb nj\u00ebjt\u00ebs skem\u00eb t\u00eb vjet\u00ebr - replikimi master-slave, pastaj kalimi, ne do ta arrijm\u00eb replik\u00ebn dhe do t\u00eb jetojm\u00eb mir\u00eb. Por n\u00ebse do t\u00eb mbajm\u00eb DB-n\u00eb ton\u00eb brenda klasterit, at\u00ebher\u00eb n\u00eb parim ka shum\u00eb zgjidhje t\u00eb gatshme p\u00ebr organizimin e t\u00eb nj\u00ebjt\u00ebs replik\u00eb master-slave, shum\u00eb zgjidhje p\u00ebr ngritjen e DB-s\u00eb brenda kuberit.
    \n P\u00ebr rezervimin e bazave t\u00eb t\u00eb dh\u00ebnave jan\u00eb lexuar nj\u00eb miliard raporte, jan\u00eb shkruar nj\u00eb miliard artikuj, nuk ka asgj\u00eb t\u00eb re k\u00ebtu, n\u00eb thelb. N\u00eb p\u00ebrgjith\u00ebsi, ndiqni \u00ebndrr\u00ebn tuaj, jetoni si t\u00eb doni, shpikni ndonj\u00eb p\u00ebrfitim t\u00eb nd\u00ebrlikuar, por patjet\u00ebr mendoni p\u00ebr m\u00ebnyr\u00ebn se si do t\u00eb rezervoni gjith\u00e7ka k\u00ebt\u00eb.<\/p>\n

    Tani tani, si si do t\u00eb ndodh\u00eb procesi i kalimit n\u00eb platform\u00ebn rezerv\u00eb n\u00eb rast zjarri. N\u00eb radh\u00eb t\u00eb par\u00eb, deploym stateless-aplikacionet paralelisht. Ato nuk ndikojn\u00eb n\u00eb logjik\u00ebn e biznesit t\u00eb aplikacioneve tona, t\u00eb projektit ton\u00eb; ne mund t\u00eb mbajm\u00eb vazhdimisht dy grupe t\u00eb hapura aplikacionesh, dhe ato mund t\u00eb fillojn\u00eb t\u00eb pranojn\u00eb trafik. \u00cbsht\u00eb shum\u00eb e r\u00ebnd\u00ebsishme q\u00eb, gjat\u00eb procesit t\u00eb kalimit n\u00eb platform\u00ebn rezerv\u00eb, t\u00eb shohim n\u00ebse nevojitet t\u00eb ri-konfigurojm\u00eb? P\u00ebr shembull, kemi klasterin e prodhimit Kubernetes, nj\u00eb klaster rezerv\u00eb Kubernetes, nj\u00eb baz\u00eb t\u00eb dh\u00ebnash master t\u00eb jashtme dhe nj\u00eb baz\u00eb t\u00eb dh\u00ebnash master rezerv\u00eb. Kemi kat\u00ebr opsione se si k\u00ebto aplikacione n\u00eb prodhim mund t\u00eb fillojn\u00eb t\u00eb nd\u00ebrveprojn\u00eb mes tyre. Mund t\u00eb kalojm\u00eb n\u00eb nj\u00eb baz\u00eb, dhe mund t\u00eb ndodh\u00eb q\u00eb duhet t\u00eb kalojm\u00eb trafik n\u00eb klasterin e prodhimit n\u00eb baz\u00ebn e re, ose mund t\u00eb ndodh\u00eb q\u00eb klasteri t\u00eb d\u00ebshtoj\u00eb \u2014 dhe kalojm\u00eb n\u00eb rezerv\u00eb, por vazhdojm\u00eb t\u00eb punojm\u00eb me baz\u00ebn e prodhimit, dhe ka nj\u00eb variant t\u00eb tret\u00eb, kur e njejta ndodh me nj\u00ebra ose tjetr\u00ebn, dhe ne kalojm\u00eb t\u00eb dy aplikacionet, kemi ri-konfiguruar rregullat tona, p\u00ebr t\u00eb siguruar q\u00eb aplikacionet e reja t\u00eb punojn\u00eb me baz\u00ebn e re t\u00eb t\u00eb dh\u00ebnave. <\/p>\n

    Pra, \u00e7far\u00eb p\u00ebrfundimesh mund t\u00eb nxjerrim nga e gjith\u00eb kjo? <\/p>\n

    \"Rezervimi
    \n
    \nP\u00ebrfundimi i par\u00eb: t\u00eb jetosh me rezerv\u00ebn \u00ebsht\u00eb mir\u00eb. Por \u00ebsht\u00eb e shtrenjt\u00eb. Idealisht, nuk duhet t\u00eb kesh vet\u00ebm nj\u00eb rezerv\u00eb. Idealisht, duhet t\u00eb kesh disa rezerva. N\u00eb radh\u00eb t\u00eb par\u00eb, rezervat duhet t\u00eb jen\u00eb minimumi, jo n\u00eb nj\u00eb qend\u00ebr t\u00eb dh\u00ebnash, dhe, n\u00eb radh\u00eb t\u00eb dyt\u00eb, minimumi, te nj\u00eb hoster tjet\u00ebr. Ka ndodhur shpesh \u2014 dhe kjo \u00ebsht\u00eb p\u00ebrvoja ime. Nuk mund t\u00eb citoj projekte, por p\u00ebr fat t\u00eb keq, ndodhi nj\u00eb zjarr n\u00eb nj\u00eb qend\u00ebr t\u00eb dh\u00ebnash... Un\u00eb thash\u00eb: kalojm\u00eb n\u00eb rezerv\u00eb! Por server\u00ebt rezerv\u00eb ndodheshin po n\u00eb t\u00eb nj\u00ebjtin raft...<\/p>\n

    Ose imagjinoni se Amazon u ndalua n\u00eb Rusi (dhe kjo ka ndodhur). Dhe gjith\u00e7ka: \u00e7far\u00eb dobie ka q\u00eb rezervat tona ndodhen n\u00eb nj\u00eb Amazon tjet\u00ebr? Ai \u00ebsht\u00eb gjithashtu i pap\u00ebrdorsh\u00ebm. Pra, e them s\u00ebrish: mbajm\u00eb rezervat, minimumi, n\u00eb nj\u00eb qend\u00ebr t\u00eb dh\u00ebnash tjet\u00ebr, dhe sa m\u00eb mir\u00eb \u2014 te nj\u00eb hoster tjet\u00ebr. <\/p>\n

    Pika e dyt\u00eb: n\u00ebse aplikacioni juaj n\u00eb Kubernetes komunikon me ndonj\u00eb burim t\u00eb jasht\u00ebm (mund t\u00eb jet\u00eb nj\u00eb baz\u00eb t\u00eb dh\u00ebnash, ashtu si\u00e7 mund t\u00eb jet\u00eb ndonj\u00eb api e jashtme), sigurohuni ta p\u00ebrcaktoni at\u00eb si nj\u00eb sh\u00ebrbim me nj\u00eb Endpoint t\u00eb jasht\u00ebm, q\u00eb n\u00eb momentin e kalimit t\u00eb mos e rindeployoni 15 aplikacionet tuaja q\u00eb po komunikojn\u00eb me t\u00eb nj\u00ebjt\u00ebn baz\u00eb. P\u00ebrcaktoni baz\u00ebn si nj\u00eb sh\u00ebrbim t\u00eb ve\u00e7ant\u00eb, lidheni me t\u00eb sikur ajo t\u00eb ishte brenda klasterit tuaj: n\u00ebse baz\u00ebn e humbni, do t\u00eb ndryshoni ip-n\u00eb n\u00eb nj\u00eb vend dhe vazhdoni t\u00eb jetoni t\u00eb lumtur. <\/p>\n

    Dhe p\u00ebrfundimisht: un\u00eb e dashuroj \"kubikun\", ashtu si edhe eksperimentet me t\u00eb. Po ashtu m\u00eb p\u00eblqen t\u00eb ndaja rezultatet e k\u00ebtyre eksperiment\u00ebve dhe p\u00ebrvoj\u00ebn time personale. Prandaj regjistrova nj\u00eb seri webinar\u00ebsh mbi K8s, mir\u00eb se vini n\u00eb kanalin ton\u00eb youtube<\/a><\/noindex> p\u00ebr detaje.
    \n
    Burimi:     habr.com<\/a><\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"

    \u041c\u0435\u043d\u044f \u0437\u043e\u0432\u0443\u0442 \u0421\u0435\u0440\u0433\u0435\u0439, \u044f \u0438\u0437 \u043a\u043e\u043c\u043f\u0430\u043d\u0438\u0438 ITSumma, \u0438 \u044f \u0445\u043e\u0447\u0443 \u0432\u0430\u043c \u0440\u0430\u0441\u0441\u043a\u0430\u0437\u0430\u0442\u044c, \u043a\u0430\u043a \u043c\u044b \u043f\u043e\u0434\u0445\u043e\u0434\u0438\u043c \u043a \u0440\u0435\u0437\u0435\u0440\u0432\u0438\u0440\u043e\u0432\u0430\u043d\u0438\u044e \u0432 Kubernetes. \u0412 \u043f\u043e\u0441\u043b\u0435\u0434\u043d\u0435\u0435 \u0432\u0440\u0435\u043c\u044f \u044f \u043c\u043d\u043e\u0433\u043e \u0437\u0430\u043d\u0438\u043c\u0430\u044e\u0441\u044c \u043a\u043e\u043d\u0441\u0443\u043b\u044c\u0442\u0430\u0442\u0438\u0432\u043d\u043e\u0439 \u0440\u0430\u0431\u043e\u0442\u043e\u0439 \u043f\u043e \u0432\u043d\u0435\u0434\u0440\u0435\u043d\u0438\u044e \u0440\u0430\u0437\u043d\u043e\u043e\u0431\u0440\u0430\u0437\u043d\u044b\u0445 devops-\u0440\u0435\u0448\u0435\u043d\u0438\u0439 \u0434\u043b\u044f \u0440\u0430\u0437\u043b\u0438\u0447\u043d\u044b\u0445 \u043a\u043e\u043c\u0430\u043d\u0434, \u0438, \u0432 \u0447\u0430\u0441\u0442\u043d\u043e\u0441\u0442\u0438, \u043f\u043b\u043e\u0442\u043d\u043e \u0440\u0430\u0431\u043e\u0442\u0430\u044e \u043f\u043e \u043f\u0440\u043e\u0435\u043a\u0442\u0430\u043c \u0441 \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u043d\u0438\u0435\u043c K8s. \u041d\u0430 \u043a\u043e\u043d\u0444\u0435\u0440\u0435\u043d\u0446\u0438\u0438 Uptime day 4, \u043a\u043e\u0442\u043e\u0440\u0430\u044f \u0431\u044b\u043b\u0430 \u043f\u043e\u0441\u0432\u044f\u0449\u0435\u043d\u0430 \u0440\u0435\u0437\u0435\u0440\u0432\u0438\u0440\u043e\u0432\u0430\u043d\u0438\u044e \u0432 \u0441\u043b\u043e\u0436\u043d\u044b\u0445 […]<\/p>\n","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"author":1,"featured_media":25449,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[688],"tags":[],"class_list":["post-33773","post","type-post","status-publish","format-standard","has-post-thumbnail","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\udd47Rezervimi n\u00eb Kubernetes: ai ekziston | ProHoster","description":"Emri im \u00ebsht\u00eb Sergey, jam nga kompania ITSumma dhe d\u00ebshiroj t'ju tregoj se si i qasje k\u00ebsaj \u00e7\u00ebshtjeje t\u00eb rezervimit n\u00eb Kubernetes.","canonical_url":"https:\/\/prohoster.info\/sq\/blog\/administrirovanie\/rezervirovanie-v-kubernetes-ono-sushhestvuet","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"sq_AL","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\u0435\u0437\u0435\u0440\u0432\u0438\u0440\u043e\u0432\u0430\u043d\u0438\u0435 \u0432 Kubernetes: \u043e\u043d\u043e \u0441\u0443\u0449\u0435\u0441\u0442\u0432\u0443\u0435\u0442 | ProHoster","og:description":"\u041c\u0435\u043d\u044f \u0437\u043e\u0432\u0443\u0442 \u0421\u0435\u0440\u0433\u0435\u0439, \u044f \u0438\u0437 \u043a\u043e\u043c\u043f\u0430\u043d\u0438\u0438 ITSumma, \u0438 \u044f \u0445\u043e\u0447\u0443 \u0432\u0430\u043c \u0440\u0430\u0441\u0441\u043a\u0430\u0437\u0430\u0442\u044c, \u043a\u0430\u043a \u043c\u044b \u043f\u043e\u0434\u0445\u043e\u0434\u0438\u043c \u043a \u0440\u0435\u0437\u0435\u0440\u0432\u0438\u0440\u043e\u0432\u0430\u043d\u0438\u044e \u0432 Kubernetes.","og:url":"https:\/\/prohoster.info\/sq\/blog\/administrirovanie\/rezervirovanie-v-kubernetes-ono-sushhestvuet","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":"2019-10-31T18:54:36+00:00","article:modified_time":"2019-10-31T18:54:36+00:00","article:publisher":"https:\/\/www.facebook.com\/prohoster","article:author":"https:\/\/www.facebook.com\/prohoster"},"aioseo_meta_data":{"post_id":"33773","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 17:03:20","breadcrumb_settings":null,"limit_modified_date":false,"reviewed_by":null,"ai":null,"created":"2021-03-01 02:33:25","updated":"2026-02-09 17:03:20","focus_keyword":null,"additional_keywords":null,"truseo_locale":null},"gt_translate_keys":[{"key":"link","format":"url"}],"_links":{"self":[{"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/posts\/33773","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/comments?post=33773"}],"version-history":[{"count":1,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/posts\/33773\/revisions"}],"predecessor-version":[{"id":159074,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/posts\/33773\/revisions\/159074"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/media\/25449"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/media?parent=33773"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/categories?post=33773"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/tags?post=33773"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}