Hi,
the cassandra cpp driver I'm using to connect to a cluster prints the
following warning when I'm creating a keyspace from my application:
1516956724.720 [WARN] (src/token_map.cpp:396:void
cass::ReplicationStrategy<Partitioner>::build_replicas_network_topology(const
TokenHostVec&, const cass::DatacenterMap&,
cass::ReplicationStrategy<Partitioner>::TokenReplicasVec&) const [with
Partitioner = cass::Murmur3Partitioner;
cass::ReplicationStrategy<Partitioner>::TokenHostVec =
std::vector<std::pair<long int, cass::Host*>,
std::allocator<std::pair<long int, cass::Host*> > >; typename
Partitioner::Token = long int;
cass::ReplicationStrategy<Partitioner>::TokenReplicasVec =
std::vector<std::pair<long int,
cass::CopyOnWritePtr<std::vector<cass::SharedRefPtr<cass::Host> > > >,
std::allocator<std::pair<long int,
cass::CopyOnWritePtr<std::vector<cass::SharedRefPtr<cass::Host> > > >
> >]): No nodes in datacenter 'DC-Ffm-2'. Check your replication
strategies.
with this statement:
CREATE KEYSPACE IF NOT EXISTS infra WITH replication = { 'class':
'NetworkTopologyStrategy', 'DC-Ffm-1': 3, 'DC-Ffm-2': 3 };
I have setup a cluster with 2 data centers each with 3 nodes:
$ nodetool status
Datacenter: DC-Ffm-1
====================
Status=Up/Down
|/ State=Normal/Leaving/Joining/Moving
-- Address Load Tokens Owns (effective) Host ID
Rack
UN 10.84.243.175 367.11 KiB 256 66.1%
23cf2ad0-faa8-41eb-99fa-fec155cf2f6e RAC-3
UN 10.84.243.165 388.85 KiB 256 69.2%
21caf970-7490-4995-b777-de0851138fb1 RAC-1
UN 10.84.243.166 337.06 KiB 256 64.7%
d2c48c30-5531-4d23-bbb9-b249cbf1816c RAC-2
Datacenter: DC-Ffm-2
====================
Status=Up/Down
|/ State=Normal/Leaving/Joining/Moving
-- Address Load Tokens Owns (effective) Host ID
Rack
UN 10.84.243.173 332.23 KiB 256 70.9%
e92617cc-e67f-4377-ab5e-d147ceb7c770 RAC-1
UN 10.84.243.174 316.64 KiB 256 62.4%
2b503a16-85ad-4273-a3a3-d476500d5556 RAC-2
UN 10.84.243.176 306.06 KiB 256 66.7%
03ff6033-b58a-45f4-a56b-1b03ed84c742 RAC-3
I'm using "endpoint_snitch: GossipingPropertyFileSnitch"
Since I have nodes in datacenter DC-Ffm-2 I don't understand the
warning message.
I want to have 3 copies of my data in each data center. I'm not sure
if the configuration of my cluster is correct or if there's a bug in
my application, in the cpp driver, ...
I'm using cassandra-cpp-driver-2.8.0-1.el7.centos.x86_64.rpm and
cassandra-3.11.1-1.noarch.rpm on red hat 7 server.
Regards
Oliver
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
