Repairing a cluster with HBCK2
HBCK2 is the repair tool for Apache HBase clusters.
It is included in the hbase-operator-tools.
The hbase-operator-tools are installed in the Docker image provided by Stackable in the directory /stackable/hbase-operator-tools/
.
HBCK2 can be executed via the wrapper script hbck2
.
For instance, if you followed the First steps then you have a cluster called simple-hbase
running.
Let us assume that the HBase user interface reports unknown RegionServers, then you can schedule the ServerCrashProcedure for these RegionServers by applying the following Job:
---
apiVersion: batch/v1
kind: Job
metadata:
name: hbck2-recover-unknown
spec:
template:
spec:
containers:
- name: hbck2
image: docker.stackable.tech/stackable/hbase:2.4.17-stackable0.0.0-dev
volumeMounts:
- name: hbase-config
mountPath: /stackable/conf
env:
- name: HBASE_CONF_DIR
value: /stackable/conf
- name: HADOOP_CONF_DIR
value: /stackable/conf
command:
- hbck2
args:
- recoverUnknown
volumes:
- name: hbase-config
projected:
sources:
- configMap:
name: simple-hdfs
- configMap:
name: simple-hbase-master-default