Description of problem: Cores are getting generated in one of the gluster pod while performing heketi device resync operation.


Version-Release number of selected component (if applicable):
6.0.0-7.4.el7rhgs.x86_64


How reproducible:
1/1

Steps to Reproduce:
1. Create and delete pvcs in while loop.
while true
do
    for i in {1..50}
    do
        ./pvc_create.sh c$i 2
        sleep 10
    done

    sleep 20

    for i in {1..50}
    do
        oc delete pvc c$i;
        sleep 10
    done
done

2. After some time when there is mismatch in between gluster vol list and heketi vol list perform heketi-cli device resync operation
# heketi-cli device resync device_id

Actual results:


Expected results:


Additional info:

---------pvc_create.sh---------------------------------------------------------

echo "kind: PersistentVolumeClaim" > glusterfs-pvc-claim.yaml
echo "apiVersion: v1" >> glusterfs-pvc-claim.yaml
echo "metadata:" >> glusterfs-pvc-claim.yaml
echo "  name: "$1 >> glusterfs-pvc-claim.yaml
echo "  annotations:" >> glusterfs-pvc-claim.yaml
echo "    volume.beta.kubernetes.io/storage-class: container" >> glusterfs-pvc-claim.yaml
echo "spec:" >> glusterfs-pvc-claim.yaml
echo "  accessModes:" >> glusterfs-pvc-claim.yaml
echo "   - ReadWriteOnce" >> glusterfs-pvc-claim.yaml
echo "  resources:" >> glusterfs-pvc-claim.yaml
echo "    requests:" >> glusterfs-pvc-claim.yaml
echo "      storage: "$2"Gi" >> glusterfs-pvc-claim.yaml

oc create -f glusterfs-pvc-claim.yaml

-----------------END-----------------------------------------------------------