StatefulSet replicas on a node has shared memory - kubernetes

Question

I am trying to deploy a stateful set of IPFS replicas on three Kubernetes worker nodes (based on this repo). The first three replicas work properly, but when it comes to the fourth one, it appears that the persistentVolumeClaims point to the shared physical memory. Therefore, the fourth node cannot acquire the lock. What would be the standard way to deploy many IPFS replicas in Kubernetes?
The fourth node printed the following log:
08:44:19.785 DEBUG cmd/ipfs: config path is /data/ipfs main.go:257
08:44:19.785 INFO cmd/ipfs: IPFS_PATH /data/ipfs main.go:301
08:44:19.785 DEBUG cmd/ipfs: Command cannot run on daemon. Checking if daemon is locked main.go:434
08:44:19.785 DEBUG lock: Checking lock lock.go:32
08:44:19.785 DEBUG lock: Can't lock file: /data/ipfs/repo.lock.
reason: cannot acquire lock: Lock FcntlFlock of /data/ipfs/repo.lock failed: resource temporarily unavailable lock.go:44
08:44:19.785 DEBUG fsrepo: (true)<->Lock is held at /data/ipfs fsrepo.go:302
Error: ipfs daemon is running. please stop it to run this command
Use 'ipfs daemon --help' for information about this command
Here is the yaml file for the stateful set:
apiVersion: apps/v1
kind: StatefulSet
metadata:
name: ipfs
namespace: ipfs
spec:
selector:
matchLabels:
app: ipfs
serviceName: ipfs
replicas: 6
template:
metadata:
labels:
app: ipfs
spec:
initContainers:
- name: init-repo
image: ipfs/go-ipfs:v0.4.11#sha256:e977e1560b960933061efc694c937d711ce1a51aa4a5239acfdff01504b11054
env:
- name: NODE_NAME
valueFrom:
fieldRef:
fieldPath: spec.nodeName
command: ['/bin/sh', '/etc/ipfs-config/init.sh']
volumeMounts:
- name: www
mountPath: /data/ipfs
- name: secrets
mountPath: /etc/ipfs-secrets
- name: config
mountPath: /etc/ipfs-config
- name: init-peers
image: ipfs/go-ipfs:v0.4.11#sha256:e977e1560b960933061efc694c937d711ce1a51aa4a5239acfdff01504b11054
command: ['/bin/sh', '/etc/ipfs-config/peers-kubernetes-refresh.sh']
volumeMounts:
- name: www
mountPath: /data/ipfs
- name: config
mountPath: /etc/ipfs-config
containers:
- name: ipfs
image: ipfs/go-ipfs:v0.4.11#sha256:e977e1560b960933061efc694c937d711ce1a51aa4a5239acfdff01504b11054
env:
- name: IPFS_LOGGING
value: debug
command:
- ipfs
- daemon
ports:
- containerPort: 4001
name: swarm
- containerPort: 5001
name: api
- containerPort: 8080
name: readonly
volumeMounts:
- name: www
mountPath: /data/ipfs
volumes:
- name: secrets
secret:
secretName: ipfs
- name: config
configMap:
name: ipfs-config
- name: www
persistentVolumeClaim:
claimName: ipfs-pvc
Here is the persistent volume definition
apiVersion: v1
kind: PersistentVolume
metadata:
name: ipfs-pv
namespace: ipfs
labels:
type: local
spec:
storageClassName: manual
capacity:
storage: 200Mi
accessModes:
- ReadWriteOnce
hostPath:
path: "/mnt/data"
And the persistent volume claim definition:
vapiVersion: v1
kind: PersistentVolumeClaim
metadata:
name: ipfs-pvc
namespace: ipfs
spec:
storageClassName: manual
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 200Mi
kubectl describe of the failing node is as follows:
Name: ipfs-3
Namespace: ipfs
Priority: 0
Node: swift-153/10.70.20.153
Start Time: Tue, 27 Oct 2020 14:38:11 -0400
Labels: app=ipfs
controller-revision-hash=ipfs-74bb88dbb6
statefulset.kubernetes.io/pod-name=ipfs-3
Annotations: <none>
Status: Running
IP: 10.244.3.43
IPs:
IP: 10.244.3.43
Controlled By: StatefulSet/ipfs
Containers:
ipfs:
Container ID: docker://81349e969be9ffcafeb4d65adf9d0b2de7311e46068e36dd4f227f169f6dfcab
Image: ipfs/go-ipfs:v0.4.11#sha256:e977e1560b960933061efc694c937d711ce1a51aa4a5239acfdff01504b11054
Image ID: docker-pullable://ipfs/go-ipfs#sha256:e977e1560b960933061efc694c937d711ce1a51aa4a5239acfdff01504b11054
Ports: 4001/TCP, 5001/TCP, 8080/TCP
Host Ports: 0/TCP, 0/TCP, 0/TCP
Command:
ipfs
daemon
State: Waiting
Reason: CrashLoopBackOff
Last State: Terminated
Reason: Error
Exit Code: 1
Started: Tue, 27 Oct 2020 14:39:51 -0400
Finished: Tue, 27 Oct 2020 14:39:51 -0400
Ready: False
Restart Count: 4
Environment:
IPFS_LOGGING: debug
Mounts:
/data/ipfs from www (rw)
/var/run/secrets/kubernetes.io/serviceaccount from default-token-hb785 (ro)
Conditions:
Type Status
Initialized True
Ready False
ContainersReady False
PodScheduled True
Volumes:
secrets:
Type: Secret (a volume populated by a Secret)
SecretName: ipfs
Optional: false
config:
Type: ConfigMap (a volume populated by a ConfigMap)
Name: ipfs-config
Optional: false
www:
Type: PersistentVolumeClaim (a reference to a PersistentVolumeClaim in the same namespace)
ClaimName: ipfs-pvc
ReadOnly: false
default-token-hb785:
Type: Secret (a volume populated by a Secret)
SecretName: default-token-hb785
Optional: false
QoS Class: BestEffort
Node-Selectors: <none>
Tolerations: node.kubernetes.io/not-ready:NoExecute op=Exists for 300s
node.kubernetes.io/unreachable:NoExecute op=Exists for 300s
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Scheduled 2m24s default-scheduler Successfully assigned ipfs/ipfs-3 to swift-153
Normal Pulled 2m2s (x3 over 2m21s) kubelet Container image "ipfs/go-ipfs:v0.4.11#sha256:e977e1560b960933061efc694c937d711ce1a51aa4a5239acfdff01504b11054" already present on machine
Normal Created 2m (x3 over 2m19s) kubelet Created container ipfs
Normal Started 2m (x3 over 2m19s) kubelet Started container ipfs
Warning DNSConfigForming 103s (x10 over 2m24s) kubelet Search Line limits were exceeded, some search paths have been omitted, the applied search line is: ipfs.svc.cluster.local svc.cluster.local cluster.local search syslab.sandbox cs.toronto.edu
Warning BackOff 103s (x6 over 2m15s) kubelet Back-off restarting failed container