igomez06
commented
8 years ago This is actually is not a bug. The cluster version is a means for commands that don't have the same cluster definition to be rejected. This a means of protection if you will. It's up to the script user to update the client, by either doing:
c = Client('ip_address', cluster_version=<>)
or
c.cluster_version=<>
Once clusterversion changed by Client, cluster version didn't update in client side, results in subsequent operation's cluster version missmatch failure. Client code: c=Client("172.20.1.95",8123) c.connect() c.setClusterVersion(2) print c.get('k')
Traceback (most recent call last): File "cluster_client.py", line 8, in
print c.get('k')
File "/root/GitHub/kinetic-py-fork/kinetic/deprecated/blockingclient.py", line 63, in get
return self._process(operations.Get(), _args, _kwargs)
File "/root/GitHub/kinetic-py-fork/kinetic/baseasync.py", line 165, in _process
return super(BaseAsync, self)._process(op, _args, _kwargs)
File "/root/GitHub/kinetic-py-fork/kinetic/deprecated/blockingclient.py", line 54, in _process
return op.onError(e)
File "/root/GitHub/kinetic-py-fork/kinetic/operations.py", line 138, in onError
raise e
kinetic.common.ClusterVersionFailureException: VERSION_FAILURE