{"id":55700,"date":"2020-01-26T00:00:00","date_gmt":"2020-01-25T21:00:00","guid":{"rendered":"https:\/\/prohoster.info\/blog\/blog_prohoster\/trassirovka-i-monitoring-v-istio-mikroservisy-i-printsip-neopredelennosti"},"modified":"2020-02-18T14:03:50","modified_gmt":"2020-02-18T11:03:50","slug":"trassirovka-i-monitoring-v-istio-mikroservisy-i-printsip-neopredelennosti","status":"publish","type":"post","link":"https:\/\/prohoster.info\/et\/blog\/administrirovanie\/trassirovka-i-monitoring-v-istio-mikroservisy-i-printsip-neopredelennosti","title":{"rendered":"Traceroute ja j\u00e4lgimine Istios: mikroteenused ja ebakindluse p\u00f5him\u00f5te","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

Heisenbergi ebamugavuse printsiip \u00fctleb, et ei saa samaaegselt m\u00f5\u00f5ta objekti asukohta ja selle kiirus. Kui objekt liigub, pole tal asukohta. Kui asukoht on olemas \u2013 t\u00e4hendab, et tal pole kiirus.<\/p>\n

\"Traceroute
\n
\nMis puudutab mikroteenuseid Red Hat OpenShifti (ja Kubernetes'i) platvormil, siis t\u00e4nu sobivale avatud l\u00e4htekoodiga tarkvarale saavad nad samal ajal edastada teavet oma j\u00f5udluse ja toimivuse kohta. See ei vaidlusta vanameest Heisenbergi, kuid kaotab ebamugavuse pilve rakenduste haldamisel. Istio v\u00f5imaldab h\u00f5lpsasti korraldada selliste rakenduste j\u00e4lgimist ja monitooringut, et k\u00f5ik p\u00fcsiks kontrolli all.
\n<\/a><\/noindex><\/p>\n

M\u00e4\u00e4ratleme terminoloogia<\/h3>\n

\nPod j\u00e4lgimine<\/b> J\u00e4lgimise (Tracing) all m\u00f5istame s\u00fcsteemi tegevuse logimist. See k\u00f5lab \u00fcsna \u00fcldiselt, kuid t\u00f5eliselt \u00fcks olulisemaid reegleid on siin visata j\u00e4lgimisandmed vastavasse salvestusruumi, ilma et oleks vaja nende vormindamisega muretseda. Ning kogu andmete otsimise ja anal\u00fc\u00fcsi t\u00f6\u00f6 lasub nende tarbija \u00f5lul. Istios kasutatakse j\u00e4lgimiss\u00fcsteemi Jaeger, mis rakendab OpenTracing'i andmemudelit.<\/p>\n

J\u00e4lgedeks<\/b> J\u00e4lgedeks (Traces, ja s\u00f5na \"j\u00e4ljed\" kasutatakse siin t\u00e4henduses \"m\u00e4rgid\", nagu n\u00e4iteks ballistilise ekspertiisi puhul) nimetame andmeid, mis kirjeldavad t\u00e4ielikult p\u00e4ringu v\u00f5i t\u00f6\u00f6\u00fclesande kulgemist, nagu \u00f6eldakse, \"algusest l\u00f5puni\". N\u00e4iteks k\u00f5ik, mis toimub hetkel, kui kasutaja vajutab nuppu veebilehel, ja kuni andmete tagastamiseni, sealhulgas k\u00f5ik kaasatud mikroteenused. \u00dcks j\u00e4lg kirjeldab t\u00e4ielikult (v\u00f5i mudeldab) p\u00e4ringu kulgemist edasi-tagasi. Jaegeri liideses jagunevad j\u00e4ljed ajakava j\u00e4rgi nagu ahel, mida saab jagada eraldi l\u00fcli. Ainult selle asemel, et l\u00fclid oleksid ahelas, koosneb j\u00e4lg nn span'idest.<\/p>\n

Span<\/b> \u2013 see on intervall t\u00f6\u00f6\u00fclesande algusest kuni selle l\u00f5puni. J\u00e4tkates analoogiat, v\u00f5ib \u00f6elda, et iga span on eraldi ahela l\u00fcli. Span'il v\u00f5ib olla (v\u00f5i mitte olla) \u00fcks v\u00f5i mitu alam-spani. Seega on k\u00f5ige \u00fclemise taseme span (root span) sama \u00fcldine kestus nagu j\u00e4lg, millele see kuulub.<\/p>\n

J\u00e4lgimine<\/b> \u2013 see, this is the actual observation of your system \u2013 through the eyes, via the UI or through automation tools. The basis of monitoring lies in tracing data. In Istio, monitoring is implemented using Prometheus and has the corresponding UI. Prometheus supports automatic monitoring with alert notifications and Alert Managers.<\/p>\n

We keep score<\/h3>\n

\nTo enable tracing, the application must create a collection of spans. They then need to be exported to Jaeger so it can create a visual representation of the trace. Among other things, these spans label the name of the operation, as well as timestamp markers for its start and end. The transmission of spans is carried out by redirecting the HTTP request headers meant for Jaeger from incoming requests to outgoing requests. Depending on the programming language used, this may require slight modifications to the application's source code. Below is an example of Java code (using the Spring Boot framework) that adds B3 (Zipkin-style) headers to your request in the Spring configuration class:<\/p>\n

\"Traceroute
\nThe following header settings are used:<\/p>\n

\"Traceroute
\nIf you are using Java, you can leave the code unchanged and simply add a few lines to the Maven POM file and set environment variables. Here are the lines to add to the POM.XML file to integrate the Jaeger Tracer Resolver:<\/p>\n

\"Traceroute
\nAnd the corresponding environment variables are set in the Dockerfile:<\/p>\n

\"Traceroute
\nThat's it, everything is set up now, and our microservices will start generating trace data.<\/p>\n

Taking a general look<\/h3>\n

\nIstio includes a simple control panel based on Grafana. Once everything is configured and running on the Red Hat OpenShift PaaS platform (in our example, Red Hat OpenShift and Kubernetes are deployed on minishift), this panel can be launched with the following command:<\/p>\n

open "$(minishift openshift service grafana -u)\/d\/1\/istio-dashboard?refresh=5⩝Id=1"\n<\/code><\/pre>\n
\nThe Grafana panel allows for a quick assessment of the system's performance. A snippet of this panel is shown in the picture below:<\/p>\n
\"Traceroute
\nSiit on n\u00e4ha, et klienditeeninduse mikroteenus kutsub v\u00e4lja eelistuste mikroteenuse versioon 1, mis omakorda kutsub v\u00e4lja soovituste mikroteenuseid versioon 1 ja 2. Grafana armatuurlaud sisaldab Dashboard Row plokki k\u00f5rgetasemeliste m\u00f5\u00f5dikute jaoks, nagu globaalne p\u00e4ringute maht, eduka p\u00e4ringu osakaal, 4xx vead. Lisaks on seal Server Mesh esitus, kus on graafikud iga teenuse jaoks ja Services Row plokk, et vaadata \u00fcksikasjalikke andmeid iga konteineri kohta igas teenuses.<\/p>\n
N\u00fc\u00fcd kaevame s\u00fcgavamale<\/h3>\n
\nKorralikult seadistatud Istio j\u00e4lgimine v\u00f5imaldab otse v\u00e4lja v\u00f5tta s\u00fcsteemi j\u00f5udluse anal\u00fc\u00fcs. Jaegeri kasutajaliideses saab j\u00e4lgimisi vaadata ja n\u00e4ha, kui kaugele ja s\u00fcgavale need ulatuvad, samuti visuaalselt tuvastada j\u00f5udluse kitsaskohti. Kui kasutada Red Hat OpenShift platvormil minishift, k\u00e4ivitatakse Jaegeri kasutajaliides j\u00e4rgmise k\u00e4suga:<\/p>\n
minishift openshift service jaeger-query --in-browser\n<\/code><\/pre>\n
\n\"Traceroute
\nMida saab sellel ekraanil j\u00e4lgimise kohta \u00f6elda:<\/p>\n