GKE: pods(dotnet application) often restart with error 139 - docker

Question

I have a private gke cluster. It contains 3 nodes (each has 2 CPUs and 7.5GB of memory) and 3 pods' replica (it's a .NET Core application). I've noticed that my containers sometimes restart with "error 139 SIGSEGV", which says that there is problem with accessing memory:
This occurs when a program attempts to access a memory location that it’s not allowed to access, or attempts to access a memory location in a way that’s not allowed.
I don't have application logs with the error before restarting the container, therefore it's impossible to debug it.
I've added a property false
in application but it didn't solve the problem.
How can I fix this problem?
Manifest:
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
name: stage-deployment
namespace: stage
spec:
replicas: 3
minReadySeconds: 5
strategy:
type: RollingUpdate
rollingUpdate:
maxSurge: 1
maxUnavailable: 0
selector:
matchLabels:
app: stage
template:
metadata:
labels:
app: stage
spec:
containers:
- name: stage-container
image: my.registry/stage/core:latest
imagePullPolicy: "Always"
ports:
- containerPort: 5000
name: http
- containerPort: 22
name: ssh
readinessProbe:
tcpSocket:
port: 5000
initialDelaySeconds: 5
periodSeconds: 10
livenessProbe:
tcpSocket:
port: 5000
initialDelaySeconds: 5
periodSeconds: 20
env:
- name: POSTGRES_DB_HOST
value: 127.0.0.1:5432
- name: POSTGRES_DB_USER
valueFrom:
secretKeyRef:
name: db-credentials
key: username
- name: POSTGRES_DB_PASSWORD
valueFrom:
secretKeyRef:
name: db-credentials
key: password
- name: DB_NAME
valueFrom:
secretKeyRef:
name: db-credentials
key: dbname
- name: cloudsql-proxy
image: gcr.io/cloudsql-docker/gce-proxy:1.11
command: ["/cloud_sql_proxy",
"-instances=my-instance:us-west1:dbserver=tcp:5432",
"-credential_file=/secrets/cloudsql/credentials.json"]
volumeMounts:
- name: instance-credentials
mountPath: /secrets/cloudsql
readOnly: true
volumes:
- name: instance-credentials
secret:
secretName: instance-credentials
imagePullSecrets:
- name: regcred
---
apiVersion: v1
kind: Service
metadata:
name: stage-service
namespace: stage
spec:
type: NodePort
selector:
app: stage
ports:
- protocol: TCP
port: 80
targetPort: 5000
name: https
---
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
annotations:
kubernetes.io/ingress.class: nginx
nginx.ingress.kubernetes.io/ssl-redirect: "true"
nginx.ingress.kubernetes.io/proxy-body-size: 300m
nginx.ingress.kubernetes.io/proxy-buffer-size: 128k
nginx.ingress.kubernetes.io/proxy-buffers-number: 4 256k
nginx.org/client-max-body-size: 1000m
name: ingress
namespace: stage
spec:
rules:
- host: my.site.com
http:
paths:
- backend:
serviceName: stage-service
servicePort: 80
tls:
- hosts:
- my.site.com
secretName: my-certs