kubernetes:stackgres:monitoring
Ez a dokumentum egy előző változata!
Tartalomjegyzék
Kubernetes StackGres
Állapot ellenőrzés, monitorozás
Kubernetes állapot ellenőrzés
A következő utasítások futtatása előtt be kell állítani a névteret.
export NAMESPACE=<névtér>
POD-ok ellenőrzése
$ kubectl get pods -n $NAMESPACE -o wide NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES pg-cluster-0 5/5 Running 6 (5d ago) 37d 10.244.0.5 kube01 <none> <none> pg-cluster-1 5/5 Running 7 (5d ago) 37d 10.244.2.5 kube03 <none> <none> pg-cluster-2 5/5 Running 8 (5d ago) 37d 10.244.1.8 kube02 <none> <none>
Amennyiben hiba van (bármelyik POD állapota nem Running, akkor az alábbi ellenőrzések szükségesek a hibás POD nevét megadva (a példában ez a pg-cluster-0 nevű pod).
$ kubectl describe pod pg-cluster-0 -n $NAMESPACE
Name: pg-cluster-0
Namespace: lapodb-postgres
Priority: 0
Service Account: pg-cluster-patroni
Node: kube01/192.168.110.161
Start Time: Sun, 03 May 2026 17:31:59 +0200
Labels: app=StackGresCluster
apps.kubernetes.io/pod-index=0
controller-revision-hash=pg-cluster-7f856d76dc
nofailover=false
noloadbalance=false
role=primary
stackgres.io/cluster=true
stackgres.io/cluster-name=pg-cluster
stackgres.io/cluster-scope=pg-cluster
stackgres.io/cluster-uid=27cdba8f-3477-4740-848d-6f82c45c405a
stackgres.io/disruptible=true
statefulset.kubernetes.io/pod-name=pg-cluster-0
Annotations: stackgres.io/cluster-controller-version: 1.18.6
stackgres.io/operatorVersion: 1.18.6
stackgres.io/patroni-version: 4.1.0
stackgres.io/pgbouncer-version: 1.25.1
stackgres.io/postgresql-version: 18.3
stackgres.io/prometheus-postgres-exporter-version: 0.19.0
status:
{"conn_url":"postgres://10.244.0.5:5432/postgres","api_url":"http://10.244.0.5:8009/patroni","state":"running","role":"primary","version":...
Status: Running
IP: 10.244.0.5
IPs:
IP: 10.244.0.5
Controlled By: StatefulSet/pg-cluster
...
Node-Selectors: node-role.kubernetes.io/postgres=true
Tolerations: node.kubernetes.io/not-ready:NoExecute op=Exists for 300s
node.kubernetes.io/unreachable:NoExecute op=Exists for 300s
Events: <none>
POD-ban található konténerek nevének meghatározása.
$ kubectl get pod pg-cluster-0 -n $NAMESPACE -o jsonpath='{.spec.containers[*].name}{"\n"}'
patroni pgbouncer prometheus-postgres-exporter postgres-util cluster-controller
A PostgreSQL példány a patroni konténerben fut.
$ kubectl logs pg-cluster-0 -c patroni -n $NAMESPACE ... 2026-06-10 08:26:26 UTC [596361]: db=postgres,user=postgres,app=[unknown],client=127.0.0.1 LOG: connection authorized: user=postgres database=postgres SSL enabled (protocol=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384, bits=256) 2026-06-10 08:26:26 UTC [596361]: db=postgres,user=postgres,app=PostgreSQL JDBC Driver,client=127.0.0.1 LOG: disconnection: session time: 0:00:00.026 user=postgres database=postgres host=127.0.0.1 port=48302 2026-06-10 08:26:33,319 INFO: no action. I am (pg-cluster-0), the leader with the lock
StackGres objektumok ellenőrzése
$ kubectl get sgcluster -n $NAMESPACE -o wide
NAME VERSION FLAVOR INSTANCES PROFILE DISK AGE REPLICATION BACKUP OBJECT STORAGE BACKUP SCHEDULE
pg-cluster 18.3 3 pg-cluster-default 8Gi 37d async
$ kubectl describe sgcluster pg-cluster -n $NAMESPACE
$ kubectl describe sgcluster pg-cluster -n $NAMESPACE
Name: pg-cluster
Namespace: lapodb-postgres
Labels: <none>
Annotations: stackgres.io/operatorVersion: 1.18.6
API Version: stackgres.io/v1
Kind: SGCluster
Metadata:
Creation Timestamp: 2026-05-03T15:25:56Z
Generation: 17
Resource Version: 8510436
UID: 27cdba8f-3477-4740-848d-6f82c45c405a
Spec:
Configurations:
Observability:
Disable Metrics: false
Prometheus Autobind: false
Sg Pooling Config: pg-cluster-default
Sg Postgres Config: pg-config
Instances: 3
Managed Sql:
Continue On SGScript Error: false
Scripts:
Id: 0
Sg Script: pg-cluster-default
Pods:
Disable Connection Pooling: false
Disable Envoy: true
Disable Metrics Exporter: false
Disable Postgres Util: false
Management Policy: OrderedReady
Persistent Volume:
Size: 8Gi
Storage Class: local-postgres
Scheduling:
Node Selector:
node-role.kubernetes.io/postgres: true
Pod Anti Affinity:
Required During Scheduling Ignored During Execution:
Label Selector:
Match Labels:
stackgres.io/cluster: pg-cluster
Topology Key: kubernetes.io/hostname
Update Strategy:
Method: InPlace
Type: OnlyDbOps
Postgres:
Flavor: vanilla
Ssl:
Enabled: true
Version: 18
Postgres Services:
Primary:
Enabled: true
Type: ClusterIP
Replicas:
Enabled: true
Type: ClusterIP
Profile: production
Replication:
Mode: async
Role: ha-read
Sg Instance Profile: pg-cluster-default
Status:
Arch: x86_64
Binding:
Name: pg-cluster-binding
Build Version: 6.49
Conditions:
Last Transition Time: 2026-05-03T15:35:52.134132Z
Reason: ClusterInitialScriptApplied
Status: True
Type: InitialScriptsApplied
Last Transition Time: 2026-05-03T15:35:52.071432Z
Reason: ClusterBootstrapped
Status: True
Type: Bootstrapped
Last Transition Time: 2026-05-03T15:31:59.887576Z
Reason: FalsePendingUpgrade
Status: False
Type: PendingUpgrade
Last Transition Time: 2026-05-03T15:31:59.887531Z
Reason: FalsePendingRestart
Status: False
Type: PendingRestart
Last Transition Time: 2026-05-03T15:31:59.313942Z
Reason: FalseFailed
Status: False
Type: Failed
Extensions:
Instances: 3
Label Selector: app=StackGresCluster,stackgres.io/cluster-name=pg-cluster,stackgres.io/cluster-scope=pg-cluster,stackgres.io/cluster=true,stackgres.io/cluster-uid=27cdba8f-3477-4740-848d-6f82c45c405a
Managed Sql:
Scripts:
Completed At: 2026-05-03T15:35:47.613868Z
Id: 0
Scripts:
Id: 0
Version: 0
Id: 4
Version: 0
Started At: 2026-05-03T15:35:44.096419Z
Updated At: 2026-05-03T15:35:47.193412Z
Os: linux
Pod Statuses:
Installed Postgres Extensions:
Name: pg-cluster-0
Node Name: kube01
Pending Restart: false
Primary: true
Replication Group: 0
Installed Postgres Extensions:
Name: pg-cluster-1
Node Name: kube03
Pending Restart: false
Primary: false
Replication Group: 0
Installed Postgres Extensions:
Name: pg-cluster-2
Node Name: kube02
Pending Restart: false
Primary: false
Replication Group: 0
Postgres Version: 18.3
Events: <none>
$ kubectl get events -n $NAMESPACE --sort-by=.lastTimestamp
LAST SEEN TYPE REASON OBJECT MESSAGE
1s Warning BackupConfigFailed sgbackup/bk2026-05-03-17-52-04 SGBackup reconciliation cycle failed: SGCluster pg-cluster has no backup configured
StatefulSet ellenőrzése
$ kubectl get statefulset -n $NAMESPACE -o wide NAME READY AGE CONTAINERS IMAGES pg-cluster 3/3 37d patroni,pgbouncer,prometheus-postgres-exporter,postgres-util,cluster-controller quay.io/ongres/patroni:v4.1.0-pg18.3-build-6.49,quay.io/ongres/pgbouncer:v1.25.1-build-6.49,quay.io/ongres/prometheus-postgres-exporter:v0.19.0-build-6.49,quay.io/ongres/postgres-util:v18.3-build-6.49,quay.io/stackgres/cluster-controller:1.18.6
PVC-k ellenőrzése
$ kubectl get pvc -n $NAMESPACE -o wide NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS VOLUMEATTRIBUTESCLASS AGE VOLUMEMODE pg-cluster-data-pg-cluster-0 Bound local-pg-kube01 8Gi RWO local-postgres <unset> 37d Filesystem pg-cluster-data-pg-cluster-1 Bound local-pg-kube03 8Gi RWO local-postgres <unset> 37d Filesystem pg-cluster-data-pg-cluster-2 Bound local-pg-kube02 8Gi RWO local-postgres <unset> 37d Filesystem
Patroni állapot ellenőrzés
Normál állapot
$ kubectl exec -n $NAMESPACE -it pg-cluster-0 -c patroni -- patronictl list + Cluster: pg-cluster (7635689612010042939) ------+----+-------------+-----+------------+-----+ | Member | Host | Role | State | TL | Receive LSN | Lag | Replay LSN | Lag | +--------------+------------+---------+-----------+----+-------------+-----+------------+-----+ | pg-cluster-0 | 10.244.0.5 | Leader | running | 2 | | | | | | pg-cluster-1 | 10.244.2.5 | Replica | streaming | 2 | 0/40002B0 | 0 | 0/40002B0 | 0 | | pg-cluster-2 | 10.244.1.8 | Replica | streaming | 2 | 0/40002B0 | 0 | 0/40002B0 | 0 | +--------------+------------+---------+-----------+----+-------------+-----+------------+-----+
kubernetes/stackgres/monitoring.1781083352.txt.gz · legutóbbi módosítás: – Szerkesztő: riba.zoltan