-
Notifications
You must be signed in to change notification settings - Fork 26
Open
Description
This has been transferred from this discussion and reflects updates since the original post.
Overview: I have recently installed a MooseFS filesystem on a cluster of servers and am having difficult starting the CSI client.
Details:
Some software versions:
- kubernetes k3s v1.28.4+k3s2
- mfsmaster version: 4.57.5-1 ; build: 1937 (now from moosefs.com ubuntu 22.04 repos)
- moosefs-csi image registry.moosefs.com:443/moosefs-csi-plugin:latest
The following csi-moosefs-config.yaml has been applied:
apiVersion: v1
kind: ConfigMap
metadata:
namespace: kube-system
name: csi-moosefs-config
data:
master_host: "server.mydomain.com" # domain name is recommended, this IP is only an example
master_port: "9421" # Fixed
k8s_root_dir: "/kubexports"
driver_working_dir: "pv_data" # Created
mount_count: "1"
mfs_logging: "true"
pv_data exists in the appropriate spot on the mfs filesystem:
root@fermat:/home/knoxuser# ls -axl /mnt/mfs/kubexports/pv_data/
total 3907
drwxr-xr-x 2 root root 1 Mar 30 14:45 .
drwxr-xr-x 18 root root 4000216 Mar 2 14:10 ..
The k3s server is started with allow-privileged=true:
ExecStart=/usr/local/bin/k3s \
server --write-kubeconfig-mode=644 --kube-apiserver-arg=allow-privileged=true
(I've tried this through the command line as above and by changing /etc/rancher/k3s/config.yaml. But I don't know how to make sure this is enabled.)
kubectl apply -f csi-moosefs.yaml leads to all pods crashing:
MooseFS-CSI# kubectl get pods | grep moosefs
csi-moosefs-controller-0 3/4 CrashLoopBackOff 4 (71s ago) 2m43s
csi-moosefs-node-wsmb9 1/2 CrashLoopBackOff 4 (74s ago) 2m43s
csi-moosefs-node-hr2b5 1/2 CrashLoopBackOff 4 (72s ago) 2m43s
csi-moosefs-node-qmrd2 1/2 CrashLoopBackOff 4 (71s ago) 2m43s
csi-moosefs-node-vr2l8 1/2 CrashLoopBackOff 4 (66s ago) 2m43s
csi-moosefs-node-pk9rw 1/2 CrashLoopBackOff 4 (67s ago) 2m43s
csi-moosefs-node-f8s2s 1/2 CrashLoopBackOff 4 (63s ago) 2m43s
Here are the details for the controller pod:
MooseFS-CSI# kubectl describe pod csi-moosefs-controller-0
Name: csi-moosefs-controller-0
Namespace: kube-system
Priority: 2000000000
Priority Class Name: system-cluster-critical
Service Account: csi-moosefs-controller-sa
Node: epyc1/10.90.1.25
Start Time: Mon, 31 Mar 2025 08:53:04 -0500
Labels: app=csi-moosefs-controller
apps.kubernetes.io/pod-index=0
controller-revision-hash=csi-moosefs-controller-6cf74f979b
role=csi-moosefs
statefulset.kubernetes.io/pod-name=csi-moosefs-controller-0
Annotations: <none>
Status: Running
IP: 10.90.1.25
IPs:
IP: 10.90.1.25
Controlled By: StatefulSet/csi-moosefs-controller
Containers:
csi-provisioner:
Container ID: containerd://7bb5d47923e112969624996cd851984978c91d6639ff1d293cae07a2eaf1c523
Image: quay.io/k8scsi/csi-provisioner:v1.4.0
Image ID: sha256:327e23baedbb939641bf5d2fb8d8ff9d8c41c2b2369cfe49a48eeef39a6e17fa
Port: <none>
Host Port: <none>
Args:
--csi-address=$(ADDRESS)
--v=5
State: Running
Started: Mon, 31 Mar 2025 08:53:05 -0500
Ready: True
Restart Count: 0
Environment:
ADDRESS: /var/lib/csi/sockets/pluginproxy/csi.sock
Mounts:
/var/lib/csi/sockets/pluginproxy/ from socket-dir (rw)
/var/run/secrets/kubernetes.io/serviceaccount from kube-api-access-fgql6 (ro)
csi-attacher:
Container ID: containerd://bcd53cfe20b2a61fbf1e0388e95a5d4c453e49f7e0e6df4ae86251642e8a1d1a
Image: quay.io/k8scsi/csi-attacher:v3.1.0
Image ID: quay.io/k8scsi/csi-attacher@sha256:e3334d505bebc44afba610e2c218ca7b4993cac61fc0951d75033ccc971a1347
Port: <none>
Host Port: <none>
Args:
--v=5
--csi-address=$(ADDRESS)
State: Running
Started: Mon, 31 Mar 2025 08:53:06 -0500
Ready: True
Restart Count: 0
Environment:
ADDRESS: /var/lib/csi/sockets/pluginproxy/csi.sock
Mounts:
/var/lib/csi/sockets/pluginproxy/ from socket-dir (rw)
/var/run/secrets/kubernetes.io/serviceaccount from kube-api-access-fgql6 (ro)
csi-resizer:
Container ID: containerd://f6619dd19e9599afff6b63d1632cf2ae6877e5d6135770fdcb8b2d13dccf1ee1
Image: quay.io/k8scsi/csi-resizer:v0.5.0
Image ID: quay.io/k8scsi/csi-resizer@sha256:6c6a0332693a7c456378f6abd2bb40611826c1e1a733cadbdae2daab3125b71c
Port: <none>
Host Port: <none>
Args:
--v=5
--csi-address=$(ADDRESS)
State: Running
Started: Mon, 31 Mar 2025 08:53:07 -0500
Ready: True
Restart Count: 0
Environment:
ADDRESS: /var/lib/csi/sockets/pluginproxy/csi.sock
Mounts:
/var/lib/csi/sockets/pluginproxy/ from socket-dir (rw)
/var/run/secrets/kubernetes.io/serviceaccount from kube-api-access-fgql6 (ro)
csi-moosefs-plugin:
Container ID: containerd://8269bdcd8a880cbe870f2ad460161cadd6c5d7b2118c4e639dde55746885f52d
Image: registry.moosefs.com:443/moosefs-csi-plugin:latest
Image ID: registry.moosefs.com:443/moosefs-csi-plugin@sha256:457bdf58227c88437c341d7ba2163dcd788da09f2e698ac88a2779f6c6dd2a05
Port: <none>
Host Port: <none>
Args:
--mode=controller
--csi-endpoint=$(CSI_ENDPOINT)
--master-host=$(MASTER_HOST)
--master-port=$(MASTER_PORT)
--root-dir=$(ROOT_DIR)
--plugin-data-dir=$(WORKING_DIR)
--mfs-logging=$(MFS_LOGGING)
State: Waiting
Reason: CrashLoopBackOff
Last State: Terminated
Reason: Error
Exit Code: 2
Started: Mon, 31 Mar 2025 09:04:10 -0500
Finished: Mon, 31 Mar 2025 09:04:10 -0500
Ready: False
Restart Count: 7
Environment:
CSI_ENDPOINT: unix:///var/lib/csi/sockets/pluginproxy/csi.sock
MASTER_HOST: <set to the key 'master_host' of config map 'csi-moosefs-config'> Optional: false
MASTER_PORT: <set to the key 'master_port' of config map 'csi-moosefs-config'> Optional: false
ROOT_DIR: <set to the key 'k8s_root_dir' of config map 'csi-moosefs-config'> Optional: false
WORKING_DIR: <set to the key 'driver_working_dir' of config map 'csi-moosefs-config'> Optional: false
MFS_LOGGING: <set to the key 'mfs_logging' of config map 'csi-moosefs-config'> Optional: false
Mounts:
/var/lib/csi/sockets/pluginproxy/ from socket-dir (rw)
/var/run/secrets/kubernetes.io/serviceaccount from kube-api-access-fgql6 (ro)
Conditions:
Type Status
Initialized True
Ready False
ContainersReady False
PodScheduled True
Volumes:
socket-dir:
Type: EmptyDir (a temporary directory that shares a pod's lifetime)
Medium:
SizeLimit: <unset>
kube-api-access-fgql6:
Type: Projected (a volume that contains injected data from multiple sources)
TokenExpirationSeconds: 3607
ConfigMapName: kube-root-ca.crt
ConfigMapOptional: <nil>
DownwardAPI: true
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 11m default-scheduler Successfully assigned kube-system/csi-moosefs-controller-0 to epyc1
Normal Pulling 11m kubelet Pulling image "quay.io/k8scsi/csi-provisioner:v1.4.0"
Normal Pulled 11m kubelet Successfully pulled image "quay.io/k8scsi/csi-provisioner:v1.4.0" in 802ms (802ms including waiting)
Normal Created 11m kubelet Created container csi-provisioner
Normal Started 11m kubelet Started container csi-provisioner
Normal Pulling 11m kubelet Pulling image "quay.io/k8scsi/csi-attacher:v3.1.0"
Normal Pulled 11m kubelet Successfully pulled image "quay.io/k8scsi/csi-attacher:v3.1.0" in 397ms (397ms including waiting)
Normal Created 11m kubelet Created container csi-attacher
Normal Started 11m kubelet Started container csi-attacher
Normal Pulling 11m kubelet Pulling image "quay.io/k8scsi/csi-resizer:v0.5.0"
Normal Pulled 11m kubelet Successfully pulled image "quay.io/k8scsi/csi-resizer:v0.5.0" in 825ms (825ms including waiting)
Normal Created 11m kubelet Created container csi-resizer
Normal Started 11m kubelet Started container csi-resizer
Normal Pulled 11m kubelet Successfully pulled image "registry.moosefs.com:443/moosefs-csi-plugin:latest" in 718ms (718ms including waiting)
Normal Pulled 11m kubelet Successfully pulled image "registry.moosefs.com:443/moosefs-csi-plugin:latest" in 613ms (613ms including waiting)
Normal Started 11m (x2 over 11m) kubelet Started container csi-moosefs-plugin
Normal Pulling 11m (x3 over 11m) kubelet Pulling image "registry.moosefs.com:443/moosefs-csi-plugin:latest"
Normal Pulled 11m kubelet Successfully pulled image "registry.moosefs.com:443/moosefs-csi-plugin:latest" in 662ms (662ms including waiting)
Normal Created 11m (x3 over 11m) kubelet Created container csi-moosefs-plugin
Warning BackOff 84s (x47 over 11m) kubelet Back-off restarting failed container csi-moosefs-plugin in pod csi-moosefs-controller-0_kube-system(1b8dfb7c-7d08-45d9-8c87-65481a07cb78)
Here are the logs on the controller pod:
MooseFS-CSI# kubectl logs csi-moosefs-controller-0
Defaulted container "csi-provisioner" out of: csi-provisioner, csi-attacher, csi-resizer, csi-moosefs-plugin
I0331 13:53:05.988702 1 feature_gate.go:216] feature gates: &{map[]}
I0331 13:53:05.988797 1 csi-provisioner.go:98] Version: v1.4.0-0-g1d9bad3
I0331 13:53:05.988825 1 csi-provisioner.go:112] Building kube configs for running in cluster...
I0331 13:53:06.020886 1 connection.go:151] Connecting to unix:///var/lib/csi/sockets/pluginproxy/csi.sock
W0331 13:53:16.021166 1 connection.go:170] Still connecting to unix:///var/lib/csi/sockets/pluginproxy/csi.sock
W0331 13:53:26.021149 1 connection.go:170] Still connecting to unix:///var/lib/csi/sockets/pluginproxy/csi.sock
The logs on the nodes:
MooseFS-CSI# kubectl logs csi-moosefs-node-f8s2s
Defaulted container "driver-registrar" out of: driver-registrar, csi-moosefs-plugin
I0331 13:53:04.862935 1 main.go:110] Version: v1.1.0-0-g80a94421
I0331 13:53:04.862973 1 main.go:120] Attempting to open a gRPC connection with: "/csi/csi.sock"
I0331 13:53:04.862999 1 connection.go:151] Connecting to unix:///csi/csi.sock
W0331 13:53:14.863248 1 connection.go:170] Still connecting to unix:///csi/csi.sock
W0331 13:53:24.863221 1 connection.go:170] Still connecting to unix:///csi/csi.sock
Metadata
Metadata
Assignees
Labels
No labels