glusterfs: failed to get the 'volume file' from server

问题

I see below error in pod logs:

, the following error information was pulled from the glusterfs log to help diagnose this issue: 
[2020-01-10 20:57:47.132637] E [glusterfsd-mgmt.c:1804:mgmt_getspec_cbk] 0-glusterfs: failed to get the 'volume file' from server
[2020-01-10 20:57:47.132690] E [glusterfsd-mgmt.c:1940:mgmt_getspec_cbk] 0-mgmt: failed to fetch volume file (key:vol_32dd7b246275)

I have glusterfs installed on three servers - server 1, 2 and 3. I am using heketi to do dynamic provisioning of PVC. PVC creation is successful but pod creation shows below status while I try to mount something on this volume:

kubectl get pods
NAME       READY   STATUS              RESTARTS   AGE
test-pod   0/2     ContainerCreating   0          4m22s

回答1:

Not an answer to Heketi's issue, but introducing an alternative solution to provide persistent storage in Kubernetes based on Gluster.

Kadalu Project(https://kadalu.io) is a Gluster based solution which natively integrates with Kubernetes without using Glusterd(Gluster's management layer).

Kadalu provides Kubernetes storage in just two steps.

Install Kadalu Operator using,

$ kubectl create -f https://kadalu.io/operator-latest.yaml

Register your storage device, through which you can provision persistent volumes to the applications running in Kubernetes.(Install Kubectl plugin in master node using pip3 install kubectl-kadalu)

$ kubectl kadalu storage-add storage-pool1 --type Replica3 \
    --device node1:/dev/vdc \
    --device node2:/dev/vdc \
    --device node3:/dev/vdc

Thats it! Now ready for PV claims. Refer this link for Quick start https://kadalu.io/docs/quick-start

Latest blog post(https://kadalu.io/blog/kadalu-kubernetes-storage) explains the different configurations available with Kadalu.

来源：https://stackoverflow.com/questions/59689007/glusterfs-failed-to-get-the-volume-file-from-server

标签

Kubernetes

GlusterFS

kubernetes-pvc