While configuring and attempting to run a 4-node / 4-host MySQL Cluster, using AWS, I've hit a roadblock while trying to register the first Data Node with the Manager.
It appears to be an issue with the fact that the data node (DN) has a public IP different from the IP associated with its eth0 (which I'll call the private ip). Thus it seems that the cluster manager receives an incoming request from the DN for NodeID:11, but the source IP address doesn't match the IP that is specified in the config.ini.
Here is the error I get on the DN while trying to connect:
2014-02-08 22:12:45 [ndbd] INFO -- Angel connected to '<mg_public_ip>:1186'
2014-02-08 22:12:48 [ndbd] INFO -- Angel allocated nodeid: 11
2014-02-08 22:12:49 [ndbd] ERROR -- Could not fetch configuration/invalid configuration, error: 'The hostname this node should have according to the configuration does not match a local interface. Attempt to bind '<dn_public_ip>' failed with error: 99 'Cannot assign requested address''
I've spent 2 days looking for solutions on-line with noluck. One poster having this problem suggested creating a network alias with _eth0 : 0_ set to the DN's public IP. Which produces the following log: (this solution also required configuration of the router's NAT).
Unable to connect with connect string: nodeid=11,<mg_public_ip>:1186
I believe this is a network configuration problem, and that _MySQL_ and _AWS_ may provide a work around, but I haven't found it yet. Any insight would be greatly appreciated!
Details:
DN Public IP == <dn_public_ip>
DN Private IP == <dn_private_ip>
The manager (MG) also has a public and private IP:
Manager Public IP == <mg_public_ip>
Manager Private IP == <mg_private_ip>
The manager is configured with the following _config.ini_:
[ndbd default]
# Options affecting ndbd processes on all data nodes:
NoOfReplicas=2 # Number of replicas
DataMemory=80M # How much memory to allocate for data storage
IndexMemory=18M # How much memory to allocate for index storage
# For DataMemory and IndexMemory, we have used the
# default values. Since the "world" database takes up
# only about 500KB, this should be more than enough for
# this example Cluster setup.
[tcp default]
# TCP/IP options:
# portnumber=2202 # This the default; however, you can use any
# port that is free for all the hosts in the cluster
# Note: It is recommended that you do not specify the port
# number at all and simply allow the default value to be used
# instead
[ndb_mgmd]
# Management process options:
hostname = <mg_private_ip> # Private IP address of MGM node
datadir=/var/lib/mysql-cluster # Directory for MGM node log files
[ndbd]
# Options for data node "A":
hostname=<dn_public_ip> # Public address of data node A
datadir=/usr/local/mysql/data # Directory for this data node's data files
NodeId=11
[ndbd]
# Options for data node "B":
hostname=*** # Hostname or IP address
datadir=/usr/local/mysql/data # Directory for this data node's data files
NodeId=12
[mysqld]
# SQL node options:
hostname=**** # Hostname or IP address
# (additional mysqld connections can be
# specified for this node for various
# purposes such as running ndb_restore)
NodeId=21
_ifconfig on Data Node A:_
eth0 Link encap:Ethernet HWaddr 22:00:0A:FE:87:82
inet <addr:dn_private_ip> Bcast:*.*.*.* Mask:255.255.255.192
inet6 addr: ***** Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:9321 errors:0 dropped:0 overruns:0 frame:0
TX packets:7950 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:739918 (722.5 KiB) TX bytes:1194550 (1.1 MiB)
Interrupt:25
_ndb_1_cluster.log on Manager:_
[MgmtSrvr] WARNING -- Found timedout nodeid reservation for 11, releasing it
[MgmtSrvr] INFO -- Nodeid 11 allocated for NDB at <dn_public_ip>