{"id":97254,"date":"2020-10-17T02:42:32","date_gmt":"2020-10-17T00:42:32","guid":{"rendered":"https:\/\/prohoster.info\/blog\/administrirovanie\/prosti-openshift-my-nedostatochno-czenili-tebya-i-prinimali-kak-dolzhnoe"},"modified":"2020-10-17T02:42:32","modified_gmt":"2020-10-17T00:42:32","slug":"prosti-openshift-my-nedostatochno-czenili-tebya-i-prinimali-kak-dolzhnoe","status":"publish","type":"post","link":"https:\/\/prohoster.info\/et\/blog\/administrirovanie\/prosti-openshift-my-nedostatochno-czenili-tebya-i-prinimali-kak-dolzhnoe","title":{"rendered":"Vabanda, OpenShift, me ei hinnanud sind piisavalt ja pidasime sind iseenesestm\u00f5istetavaks","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

See post on Kubernetes and OpenShift application development is a common topic in our client discussions.<\/p>\n

\"Vabanda,
\n
\nWe usually start by stating that Kubernetes is just Kubernetes, whereas OpenShift is a Kubernetes platform, like Microsoft AKS or Amazon EKS. Each platform has its own advantages tailored for different target audiences, and the conversation then shifts to comparing the strengths and weaknesses of specific platforms.
\n<\/a><\/noindex>
\nOverall, we considered writing this post with the conclusion that it doesn\u2019t really matter where you run the code, whether it's on OpenShift, AKS, EKS, or any custom Kubernetes\u2014let's call it KUK for brevity. \u2014it's really just the same on both.<\/i> Then we planned to take a simple 'Hello World' and show what is common and what the differences are between KUK and the Red Hat OpenShift Container Platform (hereafter referred to as OCP or simply OpenShift).<\/p>\n

However, while writing this post, we realized that we've become so accustomed to using OpenShift that we don't even recognize how it has grown into an amazing platform, far beyond just being a Kubernetes distribution. We tend to take the maturity and simplicity of OpenShift for granted, overlooking its magnificence.<\/p>\n

In summary, it\u2019s time for some active reflection, and we will step by step compare deploying our 'Hello World' on KUK and OpenShift, doing so as objectively as possible (though we might occasionally express personal views). If you're interested in strictly subjective opinions on this matter, you can read it<\/p>\n

here (EN). In this post, we will stick to facts and only facts. .<\/a><\/noindex>Clusters<\/p>\n

So, for our 'Hello World', we need clusters. Let\u2019s say 'no' to any public clouds to avoid charges for servers, registries, networks, data transfers, etc. Therefore, we choose a simple single-node cluster on<\/h3>\n

\nKUK and Minikube<\/a><\/noindex> Code Ready Containers for the OpenShift cluster. Both of these options are actually quite easy to install but will require quite a bit of resources on your laptop.<\/a><\/noindex> Building on KUK<\/p>\n

\"Vabanda,<\/p>\n

Step 1 \u2013 building our container image<\/h3>\n

\nNii et, l\u00e4hme!<\/p>\n

Let's start by deploying our 'Hello World' on minikube. For this, we will need:<\/h4>\n

\n1. Docker installed.<\/p>\n

    \n
  1. 2. Git installed.<\/li>\n
  2. 2. Paigaldatud Git.<\/li>\n
  3. 3. Paigaldatud Maven (tegelikult kasutatakse selles projektis mvnw-binaari, seega saab ka ilma selleta hakkama).<\/li>\n
  4. 4. Tegelikult l\u00e4htekood, st reposti kloon github.com\/gcolman\/quarkus-hello-world.git<\/a><\/noindex><\/li>\n<\/ol>\n

    \nEsiteks tuleb luua Quarkus'i projekt. \u00c4rge muretsege, kui te pole kunagi Quarkus.io saidiga t\u00f6\u00f6tanud \u2013 see on lihtne. Lihtsalt valige komponendid, mida soovite projektis kasutada (RestEasy, Hibernate, Amazon SQS, Camel jne), ja Quarkus seadistab maven'i arhet\u00fc\u00fcbi ja laadib k\u00f5ik github'i ilma teie sekkumiseta. See t\u00e4hendab, et s\u00f5na otseses m\u00f5ttes piisab \u00fchest hiirekl\u00f5psust \u2013 ja k\u00f5ik on valmis. Selle \u00fcle me Quarkus't armastame.<\/p>\n

    \"Vabanda,<\/p>\n

    Lihtsaim viis meie 'Hello World' konteineri pildi kokkupanemiseks on kasutada quarkus-maveni laiendusi Dockerile, mis teevad kogu vajaliku t\u00f6\u00f6 \u00e4ra. Quarkus'e tulekuga on see t\u00f5eliselt lihtne: lisage laiendus container-image-docker ja saate luua pilte maven'i k\u00e4skudega.<\/p>\n

    .\/mvnw quarkus:add-extension -Dextensions=\"container-image-docker\"\n<\/code><\/pre>\n
    \nJa l\u00f5puks, teeme meie pildi koostamise Maveniga. Tulemuseks on see, et meie l\u00e4htekood muudetakse valmis konteineripildiks, mida saab juba konteinerite k\u00e4itamisruumis k\u00e4ivitada.<\/p>\n
    \"Vabanda,<\/p>\n
    .\/mvnw -X clean package -Dquarkus.container-image.build=true\n<\/code><\/pre>\n
    \nNii et see ongi, n\u00fc\u00fcd saab konteinerit k\u00e4ivitada k\u00e4suga docker run, suunates meie teenuse porti 8080, et sellele ligi p\u00e4\u00e4seda.<\/p>\n
    docker run -i --rm -p 8080:8080 gcolman\/quarkus-hello-world\n<\/code><\/pre>\n
    \"Vabanda,<\/p>\n
    P\u00e4rast konteineri instantsi k\u00e4ivitamist j\u00e4\u00e4b vaid kontrollida k\u00e4suga curl, et meie teenus t\u00f6\u00f6tab:<\/p>\n
    \"Vabanda,<\/p>\n
    Nii et k\u00f5ik t\u00f6\u00f6tab ja see oli t\u00f5eliselt lihtne.<\/p>\n
    Samm 2 \u2013 saadame meie konteineri konteineripiltide hoidlasse<\/h4>\n
    \nPraegu hoitakse meie loodud pilti kohapeal, meie kohalikus konteinerite hoidlas. Kui tahame seda pilti kasutada oma KUBERA keskkonnas, tuleb see panna m\u00f5nda teise hoidlasse. Kubernetesis ei ole selliseid funktsioone, seega kasutame dockerhub'i. Esiteks, see on tasuta ja teiseks, (peaaegu) k\u00f5ik teevad nii.<\/p>\n
    See on samuti v\u00e4ga lihtne, ja siin on vaja ainult dockerhub'i kontot.<\/p>\n
    Nii et loome dockerhub'i ja saadame sinna meie pildi.<\/p>\n
    \"Vabanda,<\/p>\n
    Samm 3 \u2013 k\u00e4ivitame Kubernetes'e<\/h4>\n
    \nOn palju viise, kuidas koguda kubernetes'i konfiguratsiooni meie 'Hello World'i k\u00e4ivitamiseks, kuid kasutame k\u00f5ige lihtsamat, sellised me oleme...<\/p>\n
    Alustuseks k\u00e4ivitame minikube'i klastrit:<\/p>\n
    minikube start\n<\/code><\/pre>\n
    <\/p>\n
    Samm 4 \u2013 k\u00e4ivitame meie konteineripildi<\/h4>\n
    \nN\u00fc\u00fcd peame meie koodi ja konteineripildi muutma kubernetes konfigureerimiseks. Teisis\u00f5nu, vajame pod'i ja deployment definitsiooni, mis viitavad meie konteineripildile dockerhub'is. \u00dcks lihtsamaid viise seda teha on k\u00e4ivitada k\u00e4sk create deployment, viidates meie pildile:<\/p>\n
    \"Vabanda,<\/p>\n
    kubectl create deployment hello-quarkus --image=gcolman\/quarkus-hello-world:1.0.0-SNAPSHOT\n<\/code><\/pre>\n
    \nSelle k\u00e4suga palusime meie KUBEl luua deployment konfiguratsioon, mis peab sisaldama pod'i spetsifikatsiooni meie konteineripildi jaoks. See k\u00e4sk rakendab samuti seda konfiguratsiooni meie minikube klastri jaoks ja loob deployment'i, mis allalaadib meie konteineripildi ja k\u00e4ivitab pod'i klastris.<\/p>\n
    Samm 5 \u2013 avame juurdep\u00e4\u00e4su meie teenusele<\/h4>\n
    \nN\u00fc\u00fcd, kui meil on k\u00e4ivitatud konteineripilt, on aeg m\u00f5elda, kuidas konfigureerida v\u00e4lise juurdep\u00e4\u00e4su sellele Restful-teenusele, mis on tegelikult meie koodis programmeeritud.<\/p>\n
    Siin on palju viise. N\u00e4iteks saame kasutada k\u00e4sku expose, et automaatselt luua vastavad Kubernetes komponendid, nagu teenused ja l\u00f5pp-punktid. Just nii me teeme, k\u00e4ivitades k\u00e4su expose meie deployment-objekti jaoks:<\/p>\n
    kubectl expose deployment hello-quarkus --type=NodePort --port=8080\n<\/code><\/pre>\n
    \nAsetame n\u00fc\u00fcd t\u00e4helepanu expose k\u00e4su valikule \u201e--type\u201d.<\/p>\n
    Kui teeme expose ja loome komponendid, mis on vajalikud meie teenuse k\u00e4ivitamiseks, peame muu hulgas tagama, et v\u00e4ljastpoolt oleks v\u00f5imalik \u00fchendust saada teenusega hello-quarkus, mis asub meie tarkvara m\u00e4\u00e4ratud v\u00f5rgus. Ja parameeter type<\/b> v\u00f5imaldab meil luua ja \u00fchendada selliseid asju nagu koormuse tasandajad, et suunata liiklust sellesse v\u00f5rku.<\/p>\n
    N\u00e4iteks, m\u00e4\u00e4rates type=LoadBalancer<\/b>, aktiveerime automaatselt koormuse tasandaja avalikus pilves, et \u00fchendada meie Kubernetes klastri. See, muidugi, on suurep\u00e4rane, kuid tuleb m\u00f5ista, et selline konfiguratsioon on tihedalt seotud konkreetse avaliku pilvega ja seda on keerulisem \u00fcle kanda erinevate keskkondade Kubernetes'i instance'ide vahel.<\/p>\n
    Meie n\u00e4ites type=NodePort<\/b>, see, the request to our service is made via the node's IP address and port number. This option allows not to use any public clouds, but requires a number of additional steps. Firstly, you need your own load balancer, so we will deploy an NGINX load balancer in our cluster.<\/p>\n
    Step 6 \u2013 set up the load balancer<\/h4>\n
    \nMinikube has a range of platform features that simplify the creation of necessary components for external access, such as ingress controllers. Minikube comes with the Nginx ingress controller, and we just need to enable and configure it.<\/p>\n
    minikube addons enable ingress\n<\/code><\/pre>\n
    \nNow, with just one command, I will create an Nginx ingress controller that will operate inside our minikube cluster:<\/p>\n
    ingress-nginx-controller-69ccf5d9d8-j5gs9 1\/1 Running 1 33m\n<\/code><\/pre>\n
    <\/p>\n
    Step 7 \u2013 Configure ingress<\/h4>\n
    \nNow we need to configure the Nginx ingress controller so that it handles requests to hello-quarkus.<\/p>\n
    \"Vabanda,<\/p>\n
    \"Vabanda,<\/p>\n
    And finally, we need to apply this configuration.<\/p>\n
    \"Vabanda,<\/p>\n
    kubectl apply -f ingress.yml\n<\/code><\/pre>\n
    \"Vabanda,<\/p>\n
    Since we are doing all this on our own computer, we simply add the IP address of our node to the file \/etc\/hosts to direct HTTP requests to our minikube at the NGINX load balancer.<\/p>\n
    192.168.99.100 hello-quarkus.info\n<\/code><\/pre>\n
    \nThat's it, now our minikube service is available externally through the Nginx ingress controller.<\/p>\n
    \"Vabanda,<\/p>\n
    Well, that was easy, right? Or not so much?<\/h3>\n
    \"Vabanda,<\/p>\n
    Running on OpenShift (Code Ready Containers)<\/h3>\n
    \nNow let's see how this is done on the Red Hat OpenShift Container Platform (OCP).<\/p>\n
    As with minikube, we choose a single-node OpenShift cluster scheme in the form of Code Ready Containers (CRC). This used to be called minishift and was based on the OpenShift Origin project, but now it's CRC and built on the Red Hat OpenShift Container Platform.<\/p>\n
    Here we cannot help but say, 'OpenShift is wonderful!'<\/p>\n
    Initially, we thought to write that development on OpenShift is no different from development on Kubernetes. And in essence, that\u2019s true. But while writing this post, we remembered how many extra steps you have to take when you don\u2019t have OpenShift, and hence it is, we repeat, wonderful. We love when everything is easy, and the way our example is deployed and started on OpenShift compared to minikube is what actually prompted us to write this post.<\/p>\n
    Let\u2019s go through the process and see what we will need to do.<\/p>\n
    So, in the minikube example we started with Docker... Wait, we no longer need Docker installed on the machine.<\/p>\n
    Ja kohalik git pole meile vajalik.
    \nJa Maven ei ole vajalik.
    \nJa ei pea k\u00e4sitsi konteineripilti looma.
    \nJa ei pea otsima mingit konteineripiltide hoidlat.
    \nJa ingress-kontrollerit ei pea installima.
    \nJa ingress'i konfigureerimine ei ole vajalik.<\/b><\/p>\n
    Te m\u00f5istate, jah? Et meie rakendust OpenShiftis k\u00e4ivitada, ei vaja te midagi \u00fclalkirjeldatut. Ja kogu protsess n\u00e4eb v\u00e4lja j\u00e4rgmine.<\/p>\n
    Samm 1 \u2013 K\u00e4ivita oma OpenShift klaster<\/h4>\n
    \nKasutame Red Hati Code Ready Containers't, mis on tegelikult sama, mis Minikube, kuid t\u00e4isfunktsionaalne \u00fche s\u00f5lmeline OpenShifti klaster.<\/p>\n
    crc start\n<\/code><\/pre>\n
    <\/p>\n
    Samm 2 \u2013 Teostame rakenduse koondamise ja k\u00e4ivitamise OpenShifti klastris<\/h4>\n
    \nJust sellel sammul ilmneb OpenShifti lihtsus ja mugavus t\u00e4ielikult. Nagu k\u00f5igis Kubernetesi jaotustes, on meil palju viise, kuidas rakendust klastris k\u00e4ivitada. Ja nagu Kubernetesel, valime me spetsiaalselt k\u00f5ige lihtsama.<\/p>\n
    OpenShift on alati olnud loodud konteinerirakenduste loomise ja k\u00e4itamise platvormiks. Konteinerite koondamine on alati olnud selle platvormi lahutamatu osa, seet\u00f5ttu on siin palju t\u00e4iendavaid Kubernetesi ressursse vastavate \u00fclesannete jaoks.<\/p>\n
    Me kasutame OpenShifti Source 2 Image (S2I) protsessi, millel on mitu erinevat viisi, et v\u00f5tta meie l\u00e4htekood (kood v\u00f5i binaarfailid) ja muuta see konteineripildiks, mida saab k\u00e4itada OpenShiftis.<\/p>\n
    Selleks vajame kahte asja:<\/p>\n