Njengomthetho, kuhlale kunesidingo sokuhlinzeka ngemithombo yezinsiza ezizinikele esicelweni sokusebenza kwayo okulungile nokuzinzile. Kodwa kuthiwani uma izinhlelo zokusebenza ezimbalwa zisebenza ngamandla afanayo? Indlela yokuhlinzeka ngamunye wabo ngezinsiza ezincane ezidingekayo? Ungakukhawulela kanjani ukusetshenziswa kwezinsiza? Indlela yokusabalalisa kahle umthwalo phakathi kwama-node? Ungaqinisekisa kanjani ukuthi indlela yokukala evundlile iyasebenza uma umthwalo wohlelo ukhula?
Udinga ukuqala ngokuthi yiziphi izinhlobo eziyinhloko zezinsiza ezikhona ohlelweni - lokhu, yiqiniso, yisikhathi sokucubungula kanye ne-RAM. Ku-k8s kubonisa lezi zinhlobo zezinsiza zikalwa ngamayunithi alandelayo:
- I-CPU - kuma-cores
- RAM - ngamabhayithi
Ngaphezu kwalokho, ngensiza ngayinye kungenzeka ukusetha izinhlobo ezimbili zezidingo - izicelo ΠΈ imikhawulo. Izicelo - ichaza izimfuneko eziyisisekelo zezinsiza zamahhala ze-node yokusebenzisa isiqukathi (kanye ne-pod iyonke), kuyilapho imikhawulo ibeka umkhawulo oqinile ezinsizeni ezitholakalayo esitsheni.
Kubalulekile ukuqonda ukuthi i-manifest akudingeki ichaze ngokusobala zombili izinhlobo, kodwa ukuziphatha kuzoba ngale ndlela elandelayo:
- Uma kuphela imikhawulo yesisetshenziswa icaciswe ngokusobala, izicelo zalesi sisetshenziswa zithatha ngokuzenzakalelayo inani elilingana nemikhawulo (ungaqinisekisa lokhu ngokushayela izinhlangano ezichazayo). Labo. empeleni, isiqukathi sizokhawulelwa esilinganisweni esifanayo sezinsiza esizidingayo ukuze sisebenze.
- Uma kuphela izicelo ezicaciswe ngokusobala esisetshenziswa, khona-ke akukho mikhawulo ephezulu ebekiwe kulesi sisetshenziswa - i.e. isitsha sinqunyelwe kuphela izinsiza ze-node ngokwayo.
Kungenzeka futhi ukulungisa ukuphathwa kwezinsiza hhayi kuphela ezingeni lesiqukathi esithile, kodwa futhi ezingeni lendawo yegama usebenzisa lezi zinhlangano ezilandelayo:
- LimitRange β ichaza inqubomgomo yokukhawulela ezingeni lesiqukathi/i-pod ku-ns futhi iyadingeka ukuze kuchazwe imikhawulo ezenzakalelayo esitsheni/i-pod, futhi kuvinjwe ukwakhiwa kweziqukathi/ama-pods okusobala (noma okuphambene nalokho), kukhawulelwe inani lazo futhi unqume umehluko ongaba khona kumanani emikhawulo nezicelo
- I-ResourceQuotas - chaza inqubomgomo yokuvinjelwa ngokuvamile yazo zonke iziqukathi ku-ns futhi isetshenziswa, njengomthetho, ukuhlukanisa izinsiza phakathi kwezindawo (iwusizo lapho indawo inganqunywanga ngokuqinile ezingeni le-node)
Okulandelayo yizibonelo ze-manifest ezibeka imikhawulo yensiza:
-
Ezingeni elithile lesiqukathi:
containers: - name: app-nginx image: nginx resources: requests: memory: 1Gi limits: cpu: 200mLabo. kulesi simo, ukuze usebenzise isiqukathi esine-nginx, uzodinga okungenani i-1G ye-RAM yamahhala kanye no-0.2 CPU endaweni, kuyilapho kakhulu isiqukathi singadla u-0.2 CPU nayo yonke i-RAM etholakalayo endaweni.
-
Ezingeni eliphelele ns:
apiVersion: v1 kind: ResourceQuota metadata: name: nxs-test spec: hard: requests.cpu: 300m requests.memory: 1Gi limits.cpu: 700m limits.memory: 2GiLabo. isamba sazo zonke iziqukathi zesicelo eziku-ns okuzenzakalelayo azikwazi ukudlula u-300m we-CPU no-1G we-OP, futhi isamba sawo wonke umkhawulo singu-700m we-CPU kanye no-2G we-OP.
-
Imikhawulo ezenzakalelayo yeziqukathi ku-ns:
apiVersion: v1 kind: LimitRange metadata: name: nxs-limit-per-container spec: limits: - type: Container defaultRequest: cpu: 100m memory: 1Gi default: cpu: 1 memory: 2Gi min: cpu: 50m memory: 500Mi max: cpu: 2 memory: 4GiLabo. endaweni yamagama emisiwe yazo zonke iziqukathi, isicelo sizosethwa sibe ngu-100m we-CPU kanye no-1G we-OP, umkhawulo - 1 CPU no-2G. Ngesikhathi esifanayo, umkhawulo uphinda usethwe kumanani angaba khona esicelweni/umkhawulo we-CPU (50m <x <2) kanye ne-RAM (500M <x <4G).
-
Imikhawulo yezinga le-Pod ns:
apiVersion: v1 kind: LimitRange metadata: name: nxs-limit-pod spec: limits: - type: Pod max: cpu: 4 memory: 1GiLabo. ku-pod ngayinye ku-ns ezenzakalelayo kuzoba nomkhawulo ongu-4 vCPU no-1G.
Manje ngithanda ukukutshela ukuthi yiziphi izinzuzo ezingasinika le mikhawulo.
Layisha indlela yokulinganisa phakathi kwamanodi
Njengoba wazi, ingxenye ye-k8s inesibopho sokusatshalaliswa kwama-pods phakathi kwama-node, njenge umhleli, esebenza ngokuvumelana ne-algorithm ethile. Le algorithm idlula ezigabeni ezimbili lapho ukhetha i-node efanelekile okufanele uyiqalise:
- Ukuhlunga
- Iyahamba
Labo. ngokuya ngenqubomgomo echaziwe, ama-node akhethwa ekuqaleni lapho kungenzeka khona ukwethula i-pod ngokusekelwe kusethi izilandiso (kuhlanganise nokuhlola ukuthi i-node inezinsiza ezanele zokusebenzisa i-pod - PodFitsResources), bese kuba yinye yalezi zindawo, ngokusho izinto ezibalulekile amaphuzu anikezwa (okuhlanganisa, uma i-node inezinsiza zamahhala, inikezwa amaphuzu engeziwe - LeastResourceAllocation/LeastRequestedPriority/BalancedResourceAllocation) futhi i-pod yethulwa endaweni enamaphuzu amaningi (uma ama-node amaningana enelisa lesi simo ngesikhathi esisodwa, khona-ke kukhethwa okungahleliwe) .
Ngesikhathi esifanayo, udinga ukuqonda ukuthi umhleli, lapho ehlola izinsiza ezitholakalayo ze-node, iqondiswa idatha egcinwe ku-etcd - i.e. ngenani lensiza eceliwe/ekhawulelwe ye-pod ngayinye esebenza kule nodi, kodwa hhayi ukusetshenziswa kwangempela kwensiza. Lolu lwazi lungatholakala ekuphumeni komyalo kubectl describe node $NODEisibonelo:
# kubectl describe nodes nxs-k8s-s1
..
Non-terminated Pods: (9 in total)
Namespace Name CPU Requests CPU Limits Memory Requests Memory Limits AGE
--------- ---- ------------ ---------- --------------- ------------- ---
ingress-nginx nginx-ingress-controller-754b85bf44-qkt2t 0 (0%) 0 (0%) 0 (0%) 0 (0%) 233d
kube-system kube-flannel-26bl4 150m (0%) 300m (1%) 64M (0%) 500M (1%) 233d
kube-system kube-proxy-exporter-cb629 0 (0%) 0 (0%) 0 (0%) 0 (0%) 233d
kube-system kube-proxy-x9fsc 0 (0%) 0 (0%) 0 (0%) 0 (0%) 233d
kube-system nginx-proxy-k8s-worker-s1 25m (0%) 300m (1%) 32M (0%) 512M (1%) 233d
nxs-monitoring alertmanager-main-1 100m (0%) 100m (0%) 425Mi (1%) 25Mi (0%) 233d
nxs-logging filebeat-lmsmp 100m (0%) 0 (0%) 100Mi (0%) 200Mi (0%) 233d
nxs-monitoring node-exporter-v4gdq 112m (0%) 122m (0%) 200Mi (0%) 220Mi (0%) 233d
Allocated resources:
(Total limits may be over 100 percent, i.e., overcommitted.)
Resource Requests Limits
-------- -------- ------
cpu 487m (3%) 822m (5%)
memory 15856217600 (2%) 749976320 (3%)
ephemeral-storage 0 (0%) 0 (0%)Lapha sibona wonke ama-pods asebenza endaweni ethile, kanye nezinsiza ezicelwa i-pod ngayinye. Futhi nakhu ukuthi amalogi omhleli abukeka kanjani lapho i-cronjob-cron-events-1573793820-xt6q9 pod yethulwa (lolu lwazi luzovela kulogi lomhleli lapho usetha ileveli yokungena ye-10 kuma-agumenti womyalo wokuqalisa -v=10):
log
I1115 07:57:21.637791 1 scheduling_queue.go:908] About to try and schedule pod nxs-stage/cronjob-cron-events-1573793820-xt6q9
I1115 07:57:21.637804 1 scheduler.go:453] Attempting to schedule pod: nxs-stage/cronjob-cron-events-1573793820-xt6q9
I1115 07:57:21.638285 1 predicates.go:829] Schedule Pod nxs-stage/cronjob-cron-events-1573793820-xt6q9 on Node nxs-k8s-s5 is allowed, Node is running only 16 out of 110 Pods.
I1115 07:57:21.638300 1 predicates.go:829] Schedule Pod nxs-stage/cronjob-cron-events-1573793820-xt6q9 on Node nxs-k8s-s6 is allowed, Node is running only 20 out of 110 Pods.
I1115 07:57:21.638322 1 predicates.go:829] Schedule Pod nxs-stage/cronjob-cron-events-1573793820-xt6q9 on Node nxs-k8s-s3 is allowed, Node is running only 20 out of 110 Pods.
I1115 07:57:21.638322 1 predicates.go:829] Schedule Pod nxs-stage/cronjob-cron-events-1573793820-xt6q9 on Node nxs-k8s-s4 is allowed, Node is running only 17 out of 110 Pods.
I1115 07:57:21.638334 1 predicates.go:829] Schedule Pod nxs-stage/cronjob-cron-events-1573793820-xt6q9 on Node nxs-k8s-s10 is allowed, Node is running only 16 out of 110 Pods.
I1115 07:57:21.638365 1 predicates.go:829] Schedule Pod nxs-stage/cronjob-cron-events-1573793820-xt6q9 on Node nxs-k8s-s12 is allowed, Node is running only 9 out of 110 Pods.
I1115 07:57:21.638334 1 predicates.go:829] Schedule Pod nxs-stage/cronjob-cron-events-1573793820-xt6q9 on Node nxs-k8s-s11 is allowed, Node is running only 11 out of 110 Pods.
I1115 07:57:21.638385 1 predicates.go:829] Schedule Pod nxs-stage/cronjob-cron-events-1573793820-xt6q9 on Node nxs-k8s-s1 is allowed, Node is running only 19 out of 110 Pods.
I1115 07:57:21.638402 1 predicates.go:829] Schedule Pod nxs-stage/cronjob-cron-events-1573793820-xt6q9 on Node nxs-k8s-s2 is allowed, Node is running only 21 out of 110 Pods.
I1115 07:57:21.638383 1 predicates.go:829] Schedule Pod nxs-stage/cronjob-cron-events-1573793820-xt6q9 on Node nxs-k8s-s9 is allowed, Node is running only 16 out of 110 Pods.
I1115 07:57:21.638335 1 predicates.go:829] Schedule Pod nxs-stage/cronjob-cron-events-1573793820-xt6q9 on Node nxs-k8s-s8 is allowed, Node is running only 18 out of 110 Pods.
I1115 07:57:21.638408 1 predicates.go:829] Schedule Pod nxs-stage/cronjob-cron-events-1573793820-xt6q9 on Node nxs-k8s-s13 is allowed, Node is running only 8 out of 110 Pods.
I1115 07:57:21.638478 1 predicates.go:1369] Schedule Pod nxs-stage/cronjob-cron-events-1573793820-xt6q9 on Node nxs-k8s-s10 is allowed, existing pods anti-affinity terms satisfied.
I1115 07:57:21.638505 1 predicates.go:1369] Schedule Pod nxs-stage/cronjob-cron-events-1573793820-xt6q9 on Node nxs-k8s-s8 is allowed, existing pods anti-affinity terms satisfied.
I1115 07:57:21.638577 1 predicates.go:1369] Schedule Pod nxs-stage/cronjob-cron-events-1573793820-xt6q9 on Node nxs-k8s-s9 is allowed, existing pods anti-affinity terms satisfied.
I1115 07:57:21.638583 1 predicates.go:829] Schedule Pod nxs-stage/cronjob-cron-events-1573793820-xt6q9 on Node nxs-k8s-s7 is allowed, Node is running only 25 out of 110 Pods.
I1115 07:57:21.638932 1 resource_allocation.go:78] cronjob-cron-events-1573793820-xt6q9 -> nxs-k8s-s10: BalancedResourceAllocation, capacity 39900 millicores 66620178432 memory bytes, total request 2343 millicores 9640186880 memory bytes, score 9
I1115 07:57:21.638946 1 resource_allocation.go:78] cronjob-cron-events-1573793820-xt6q9 -> nxs-k8s-s10: LeastResourceAllocation, capacity 39900 millicores 66620178432 memory bytes, total request 2343 millicores 9640186880 memory bytes, score 8
I1115 07:57:21.638961 1 resource_allocation.go:78] cronjob-cron-events-1573793820-xt6q9 -> nxs-k8s-s9: BalancedResourceAllocation, capacity 39900 millicores 66620170240 memory bytes, total request 4107 millicores 11307422720 memory bytes, score 9
I1115 07:57:21.638971 1 resource_allocation.go:78] cronjob-cron-events-1573793820-xt6q9 -> nxs-k8s-s8: BalancedResourceAllocation, capacity 39900 millicores 66620178432 memory bytes, total request 5847 millicores 24333637120 memory bytes, score 7
I1115 07:57:21.638975 1 resource_allocation.go:78] cronjob-cron-events-1573793820-xt6q9 -> nxs-k8s-s9: LeastResourceAllocation, capacity 39900 millicores 66620170240 memory bytes, total request 4107 millicores 11307422720 memory bytes, score 8
I1115 07:57:21.638990 1 resource_allocation.go:78] cronjob-cron-events-1573793820-xt6q9 -> nxs-k8s-s8: LeastResourceAllocation, capacity 39900 millicores 66620178432 memory bytes, total request 5847 millicores 24333637120 memory bytes, score 7
I1115 07:57:21.639022 1 generic_scheduler.go:726] cronjob-cron-events-1573793820-xt6q9_nxs-stage -> nxs-k8s-s10: TaintTolerationPriority, Score: (10)
I1115 07:57:21.639030 1 generic_scheduler.go:726] cronjob-cron-events-1573793820-xt6q9_nxs-stage -> nxs-k8s-s8: TaintTolerationPriority, Score: (10)
I1115 07:57:21.639034 1 generic_scheduler.go:726] cronjob-cron-events-1573793820-xt6q9_nxs-stage -> nxs-k8s-s9: TaintTolerationPriority, Score: (10)
I1115 07:57:21.639041 1 generic_scheduler.go:726] cronjob-cron-events-1573793820-xt6q9_nxs-stage -> nxs-k8s-s10: NodeAffinityPriority, Score: (0)
I1115 07:57:21.639053 1 generic_scheduler.go:726] cronjob-cron-events-1573793820-xt6q9_nxs-stage -> nxs-k8s-s8: NodeAffinityPriority, Score: (0)
I1115 07:57:21.639059 1 generic_scheduler.go:726] cronjob-cron-events-1573793820-xt6q9_nxs-stage -> nxs-k8s-s9: NodeAffinityPriority, Score: (0)
I1115 07:57:21.639061 1 interpod_affinity.go:237] cronjob-cron-events-1573793820-xt6q9 -> nxs-k8s-s10: InterPodAffinityPriority, Score: (0)
I1115 07:57:21.639063 1 selector_spreading.go:146] cronjob-cron-events-1573793820-xt6q9 -> nxs-k8s-s10: SelectorSpreadPriority, Score: (10)
I1115 07:57:21.639073 1 interpod_affinity.go:237] cronjob-cron-events-1573793820-xt6q9 -> nxs-k8s-s8: InterPodAffinityPriority, Score: (0)
I1115 07:57:21.639077 1 selector_spreading.go:146] cronjob-cron-events-1573793820-xt6q9 -> nxs-k8s-s8: SelectorSpreadPriority, Score: (10)
I1115 07:57:21.639085 1 interpod_affinity.go:237] cronjob-cron-events-1573793820-xt6q9 -> nxs-k8s-s9: InterPodAffinityPriority, Score: (0)
I1115 07:57:21.639088 1 selector_spreading.go:146] cronjob-cron-events-1573793820-xt6q9 -> nxs-k8s-s9: SelectorSpreadPriority, Score: (10)
I1115 07:57:21.639103 1 generic_scheduler.go:726] cronjob-cron-events-1573793820-xt6q9_nxs-stage -> nxs-k8s-s10: SelectorSpreadPriority, Score: (10)
I1115 07:57:21.639109 1 generic_scheduler.go:726] cronjob-cron-events-1573793820-xt6q9_nxs-stage -> nxs-k8s-s8: SelectorSpreadPriority, Score: (10)
I1115 07:57:21.639114 1 generic_scheduler.go:726] cronjob-cron-events-1573793820-xt6q9_nxs-stage -> nxs-k8s-s9: SelectorSpreadPriority, Score: (10)
I1115 07:57:21.639127 1 generic_scheduler.go:781] Host nxs-k8s-s10 => Score 100037
I1115 07:57:21.639150 1 generic_scheduler.go:781] Host nxs-k8s-s8 => Score 100034
I1115 07:57:21.639154 1 generic_scheduler.go:781] Host nxs-k8s-s9 => Score 100037
I1115 07:57:21.639267 1 scheduler_binder.go:269] AssumePodVolumes for pod "nxs-stage/cronjob-cron-events-1573793820-xt6q9", node "nxs-k8s-s10"
I1115 07:57:21.639286 1 scheduler_binder.go:279] AssumePodVolumes for pod "nxs-stage/cronjob-cron-events-1573793820-xt6q9", node "nxs-k8s-s10": all PVCs bound and nothing to do
I1115 07:57:21.639333 1 factory.go:733] Attempting to bind cronjob-cron-events-1573793820-xt6q9 to nxs-k8s-s10Lapha sibona ukuthi ekuqaleni isihlungi se-schedule futhi sikhiqize uhlu lwama-node angu-3 lapho lungasetshenziswa khona (nxs-k8s-s8, nxs-k8s-s9, nxs-k8s-s10). Bese ibala amaphuzu ngokusekelwe kumapharamitha ambalwa (okuhlanganisa i-BalancedResourceAllocation, LeastResourceAllocation) ku-node ngayinye kulawa ukuze kutholwe indawo efanelekile kakhulu. Ekugcineni, i-pod ihlelwe ku-node enenani eliphakeme kakhulu lamaphuzu (lapha ama-node amabili ngesikhathi esisodwa anenani elifanayo lamaphuzu 100037, ngakho-ke kukhethwa okungahleliwe - nxs-k8s-s10).
isiphetho: uma i-node igijima ama-pods okungekho imingcele ebekiwe, khona-ke ama-k8s (kusuka endaweni yokubuka ukusetshenziswa kwezinsiza) lokhu kuzolingana njengokuthi kwakungekho ama-pods anjalo kule node nhlobo. Ngakho-ke, uma wena, ngokwemibandela, une-pod enenqubo yokuminza (isibonelo, wowza) futhi kungekho mingcele ebekiwe, khona-ke isimo singase siphakame lapho le pod empeleni idle zonke izinsiza ze-node, kodwa nge-k8s le node. ibhekwa njengengalayishiwe futhi izonikezwa inani elifanayo lamaphuzu lapho ihlelwa (ngokuqondile emaphuzwini ahlola izinsiza ezitholakalayo) njengendawo engenawo ama-pods asebenzayo, okungagcina kuholele ekusabalaliseni okungalingani komthwalo phakathi kwamanodi.
Ukukhishwa kwePod
Njengoba wazi, i-pod ngayinye inikezwe amakilasi ama-QoS ama-3:
- okuqinisekisiwe - inikezwa lapho isiqukathi ngasinye ku-pod isicelo kanye nomkhawulo kushiwo inkumbulo kanye ne-cpu, futhi la manani kufanele afane
- kuqhuma - Okungenani isitsha esisodwa kuphod sinesicelo kanye nomkhawulo, ngesicelo < umkhawulo
- umzamo omuhle kakhulu - lapho kungekho nesisodwa isiqukathi ku-pod esilinganiselwe
Ngasikhathi sinye, lapho i-node ihlangabezana nokushoda kwezinsiza (idiski, inkumbulo), i-kubelet iqala ukukala futhi ikhiphe ama-pods ngokuya nge-algorithm ethile ecabangela ukubaluleka kwe-pod nesigaba sayo se-QoS. Isibonelo, uma sikhuluma nge-RAM, bese kusekelwe ekilasini le-QoS, amaphuzu anikezwa ngokulandela isimiso esilandelayo:
- Kuqinisekisiwe:-998
- BestEffort: 1000
- Burstable: iminithi(ubukhulu(2, 1000 - (1000 * memoryRequestBytes) / machineMemoryCapacityBytes), 999)
Labo. ngokubaluleka okufanayo, i-kubelet izoqala ngokukhipha ama-pods ngomzamo omuhle kakhulu wekilasi le-QoS endaweni.
isiphetho: uma ufuna ukunciphisa amathuba okuba i-pod oyifunayo ikhishwe ku-node uma kwenzeka ukuntuleka kwezinsiza kuyo, khona-ke kanye nokubalulekile, udinga futhi ukunakekela ukusetha isicelo / umkhawulo kuso.
Indlela yokulinganisa okuzenzakalelayo kwe-application pods (HPA)
Lapho umsebenzi uwukukhulisa ngokuzenzakalelayo futhi wehlise inani lama-pods kuye ngokusetshenziswa kwezinsiza (uhlelo - CPU/RAM noma umsebenzisi - rps), i-k8s entity njenge HPA (I-Horizontal Pod Autoscaler). I-algorithm yayo ingokulandelayo:
- Ukufundwa kwamanje kwesisetshenziswa esibukiwe kuyanqunywa (currentMetricValue)
- Amanani afiswayo esisetshenziswa ayanqunywa (desiredMetricValue), okuthi ezinsizeni zesistimu asethwe kusetshenziswa isicelo
- Inombolo yamanje yezifaniso inqunywa (currentReplicas)
- Ifomula elandelayo ibala inombolo efiswayo ye-replicas (desiredReplicas)
wishdReplicas = [ currentReplicas * ( currentMetricValue / desiredMetricValue )]
Kulesi simo, ukukala ngeke kwenzeke uma i-coefficient (currentMetricValue / desiredMetricValue) iseduze no-1 (kulesi simo, singasetha iphutha elivunyelwe ngokwethu; ngokuzenzakalelayo ngu-0.1).
Ake sibheke ukuthi i-hpa isebenza kanjani sisebenzisa isibonelo sohlelo lokusebenza lokuhlola uhlelo lokusebenza (elichazwa ngokuthi Ukuthunyelwa), lapho kudingeka khona ukuguqula inani lezifaniso kuye ngokusetshenziswa kwe-CPU:
-
I-manifest yohlelo lokusebenza
kind: Deployment apiVersion: apps/v1beta2 metadata: name: app-test spec: selector: matchLabels: app: app-test replicas: 2 template: metadata: labels: app: app-test spec: containers: - name: nginx image: registry.nixys.ru/generic-images/nginx imagePullPolicy: Always resources: requests: cpu: 60m ports: - name: http containerPort: 80 - name: nginx-exporter image: nginx/nginx-prometheus-exporter resources: requests: cpu: 30m ports: - name: nginx-exporter containerPort: 9113 args: - -nginx.scrape-uri - http://127.0.0.1:80/nginx-statusLabo. sibona ukuthi i-pod yesicelo yethulwa ekuqaleni ezimweni ezimbili, ngasinye esiqukethe iziqukathi ezimbili ze-nginx kanye ne-nginx-exporter, ngayinye yazo ecacisiwe. izicelo kwe-CPU.
-
I-HPA Manifesto
apiVersion: autoscaling/v2beta2 kind: HorizontalPodAutoscaler metadata: name: app-test-hpa spec: maxReplicas: 10 minReplicas: 2 scaleTargetRef: apiVersion: extensions/v1beta1 kind: Deployment name: app-test metrics: - type: Resource resource: name: cpu target: type: Utilization averageUtilization: 30Labo. Sidale i-hpa ezoqapha ukuhlolwa kohlelo lokusebenza Lokuthunyelwa futhi ilungise inani lama-pod ngohlelo lokusebenza ngokususelwe kunkomba ye-cpu (silindele ukuthi i-pod kufanele idle u-30% we-CPU eyicelayo), nenani lezifaniso ezikhona. ububanzi 2-10.
Manje, ake sibheke indlela yokusebenza kwe-hpa uma sisebenzisa umthwalo kwenye yeziko:
# kubectl top pod NAME CPU(cores) MEMORY(bytes) app-test-78559f8f44-pgs58 101m 243Mi app-test-78559f8f44-cj4jz 4m 240Mi
Sekukonke sinokulandelayo:
- Inani elifunekayo (desiredMetricValue) - ngokuya ngezilungiselelo ze-hpa, sino-30%
- Inani lamanje (currentMetricValue) - ekubaleni, isilawuli-umphathi ubala inani elimaphakathi lokusetshenziswa kwensiza ku-%, i.e. ngokwemibandela yenza lokhu okulandelayo:
- Ithola amanani aphelele amamethrikhi e-pod kusuka kuseva yemethrikhi, i.e. 101m kanye 4m
- Ibala isilinganiso senani eliphelele, isb. (101m + 4m) / 2 = 53m
- Ithola inani eliphelele lokusetshenziswa kwensiza oyifunayo (kulokhu, izicelo zazo zonke iziqukathi zifingqwa) 60m + 30m = 90m
- Ibala iphesenti elimaphakathi lokusetshenziswa kwe-CPU ngokuhlobene ne-pod yesicelo, isb. 53m / 90m * 100% = 59%
Manje sinakho konke esikudingayo ukuze sinqume ukuthi sidinga yini ukushintsha inombolo ye-replicas; ukwenza lokhu, sibala i-coefficient:
ratio = 59% / 30% = 1.96
Labo. inani le-replicas kufanele lenyuswe ~ izikhathi ezi-2 kanye nenani libe ngu- [2 * 1.96] = 4.
Isiphetho: Njengoba ubona, ukuze lo mshini usebenze, isimo esidingekayo siwukuba khona kwezicelo zazo zonke iziqukathi ku-pod ephawuliwe.
Indlela yokulinganisa i-autoscaling yama-node (i-Cluster Autoscaler)
Ukuze kuncishiswe umthelela ongemuhle ohlelweni ngesikhathi sokukhuphuka komthwalo, ukuba ne-hpa emisiwe akwanele. Isibonelo, ngokuya ngezilungiselelo kumphathi wesilawuli se-hpa, inquma ukuthi inani le-replicas lidinga ukukhushulwa izikhathi ezi-2, kodwa ama-node awanazo izinsiza zamahhala zokusebenzisa inani elinjalo lama-pods (okungukuthi, i-node ayikwazi ukunikeza ucele izinsiza ku-pod yezicelo) futhi lawa ma-pods ashintshela kusimo esilindile.
Kulokhu, uma umhlinzeki ene-IaaS/PaaS ehambisanayo (isibonelo, i-GKE/GCE, i-AKS, i-EKS, njll.), ithuluzi elifana ne- I-Node Autoscaler. Ikuvumela ukuthi usethe inani eliphakeme nelincane lama-node ku-cluster futhi ulungise ngokuzenzakalelayo inombolo yamanje yama-node (ngokubiza i-API yomhlinzeki wamafu ukuze u-oda / ukhiphe i-node) lapho kuntuleka izinsiza kuqoqo kanye nama-pods. azikwazi ukuhlelwa (zisesimweni sokulinda).
Isiphetho: Ukuze ukwazi ukwenza ama-node okuzenzakalelayo, kuyadingeka ukusetha izicelo ezitsheni ze-pod ukuze ama-k8 akwazi ukuhlola kahle umthwalo kuma-node futhi ngokufanele abike ukuthi azikho izinsiza kuqoqo zokuqalisa i-pod elandelayo.
isiphetho
Kufanele kuqashelwe ukuthi ukusetha imikhawulo yensiza yesiqukathi akuyona imfuneko ukuze uhlelo lokusebenza lusebenze ngempumelelo, kodwa kusengcono ukwenza kanjalo ngenxa yalezi zizathu ezilandelayo:
- Ukuze uthole ukusebenza okunembe kakhudlwana kwesihleli mayelana nokulinganisa komthwalo phakathi kwama-k8s node
- Ukunciphisa amathuba okuthi kwenzeke "ukukhishwa kwe-pod".
- Ukuze i-autoscaling evundlile ye-application pods (HPA) isebenze
- Ukuze uthole ukulinganisa okuzenzakalelayo kwama-node (I-Cluster Autoscaling) kubahlinzeki bamafu
Funda nezinye izindatshana kubhulogi yethu:
Source: www.habr.com