Announcement

Announcement Module
Collapse
No announcement yet.

failed to open gcomm backend connection: 110: failed to reach primary view: 110

Page Title Module
Move Remove Collapse
X
Conversation Detail Module
Collapse
  • Filter
  • Time
  • Show
Clear All
new posts

  • failed to open gcomm backend connection: 110: failed to reach primary view: 110

    Hi,
    I established a xtradb-cluster test environment with percona-xtrabackup-2.1.7-721-Linux-x86_64.tar.gz and Percona-XtraDB-Cluster-5.6.15-25.3.706.Linux.x86_64.tar.gz
    But It does not work
    my.cnf as follows:
    ip:192.168.2.203:
    [mysqld]
    port = 3306
    socket = /tmp/mysql.sock
    datadir = /usr/local/mysql/var
    binlog_format=ROW
    default_storage_engine=InnoDB
    wsrep_cluster_address=gcomm://192.168.2.203,192.168.2.204,192.168.2.205
    wsrep_provider=/usr/local/mysql/lib/libgalera_smm.so
    wsrep_slave_threads=1
    ##wsrep_sst_donor=192.168.2.203
    wsrep_cluster_name=pxctest
    wsrep_node_address=192.168.2.203
    wsrep_sst_method=xtrabackup
    wsrep_sst_auth="sstuser:s3cret"
    innodb_autoinc_lock_mode=2


    ip:192.168.2.204
    [mysqld]
    port = 3306
    socket = /tmp/mysql.sock
    datadir = /usr/local/mysql/var
    binlog_format=ROW
    log_bin=mysql-bin
    user=mysql
    default_storage_engine=InnoDB
    wsrep_cluster_address=gcomm://192.168.2.203,192.168.2.204,192.168.2.205
    wsrep_provider=/usr/local/mysql/lib/libgalera_smm.so
    wsrep_slave_threads=1
    #wsrep_sst_donor=192.168.2.203
    wsrep_cluster_name=pxctest
    wsrep_node_address=192.168.2.204
    wsrep_sst_method=xtrabackup
    wsrep_sst_auth="sstuser:s3cret"
    #log_slave_updates
    innodb_locks_unsafe_for_binlog=1
    innodb_autoinc_lock_mode=2

    ip:192.168.2.205:
    [mysqld]
    port = 3306
    socket = /tmp/mysql.sock
    datadir = /usr/local/mysql/var
    binlog_format=ROW
    log_bin=mysql-bin
    user=mysql
    default_storage_engine=InnoDB
    wsrep_cluster_address=gcomm://192.168.2.203,192.168.2.204,192.168.2.205
    wsrep_provider=/usr/local/mysql/lib/libgalera_smm.so
    wsrep_slave_threads=1
    ##wsrep_sst_donor=192.168.2.203
    wsrep_cluster_name=pxctest
    wsrep_node_address=192.168.2.205
    wsrep_sst_method=xtrabackup
    wsrep_sst_auth="sstuser:s3cret"
    ##log_slave_updates
    innodb_locks_unsafe_for_binlog=1
    innodb_autoinc_lock_mode=2

    the error as flollows(I have disabled selinux and stopped iptables)
    2014-02-19 08:59:31 2687 [ERROR] WSREP: failed to open gcomm backend connection: 110: failed to reach primary view: 110 (Connection timed out) at gcomm/src/pc.cpp:connect():141 2014-02-19 08:59:31 2687
    [ERROR] WSREP: gcs/src/gcs_core.c:gcs_core_open():196: Failed to open backend connection: -110 (Connection timed out) 2014-02-19 08:59:31 2687
    [ERROR] WSREP: gcs/src/gcs.c:gcs_open():1291: Failed to open channel 'pxctest' at 'gcomm://192.168.2.203,192.168.2.204,192.168.2.205': -110 (Connection timed out) 2014-02-19 08:59:31 2687
    [ERROR] WSREP: gcs connect failed: Connection timed out 2014-02-19 08:59:31 2687
    [ERROR] WSREP: wsrep::connect() failed: 7 2014-02-19 08:59:31 2687
    [ERROR] Aborting 2014-02-19 08:59:31 2687 [Note] WSREP: Service disconnected. 2014-02-19 08:59:32 2687
    [Note] WSREP: Some threads may fail to exit. 2014-02-19 08:59:32 2687
    [Note] Binlog end 2014-02-19 08:59:32 2687
    [Note] /usr/local/Percona-XtraDB-Cluster-5.6.15-25.3.706.Linux.x86_64/bin/mysqld: Shutdown complete
    140219 08:59:32 mysqld_safe mysqld from pid file /usr/local/mysql/var/db1.pid ended


    I want to find a solution to solve it .

    I read the document I found:
    Prerequisites
    • All three nodes have a CentOS 6.3 installation.
    • Firewall has been set up to allow connecting to ports 3306, 4444, 4567 and 4568
    • SELinux is disabled

    Is it must be on CentOS 6.3 if I want use centos platform???

    my platform version as follows:
    [root@db3 var]# cat /etc/issue
    CentOS release 6.4 (Final)
    Kernel \r on an \m

    Does xtradb-cluster work on centos 6.4 ?
    many thanks
    Last edited by xliu; 02-19-2014, 03:25 AM.

  • #2
    Please check that you have this package installed. nc-1.84-22.e16.x86_64. Your version may be different. I am running centos 6.5. Also you may want to add this to your my.cnf
    wsrep_node_name =

    I like to call my nodes this: node0, node1, node2.

    Once you have done the above, try booting one of your nodes as a bootstrap and then once its up and running try booting a second node. Report back your progress and we can try a few more things if we need to.

    Comment


    • #3
      many thanks
      After I have installed nc-* with yum and added wsrep_node_name = node* in /etc/my.cnf in three nodes , It throws the same error after I start mysqld service


      My process(start mysqld service) as follows:
      203:
      /usr/local/mysql/bin/mysqld_safe --user=mysql wsrep_cluster_address=gcomm:// &
      204:
      /usr/local/mysql/bin/mysqld_safe --user=mysql &
      205:
      ​ /usr/local/mysql/bin/mysqld_safe --user=mysql &
      Last edited by xliu; 02-19-2014, 07:54 PM.

      Comment


      • #4
        Start 203 like you are that will be the bootstrap. Now edit you my.cnf on 204 and on this line wsrep_cluster_address=gcomm://192.168.2.203. Do not put the address of 204,or 205 just yet. If 204 connects then edit the my.cnf on 205 and do the same thing. If 205 connects, then go back to each machine and add the ip address for each node on the wsrep_cluster_address. Once you have done that restart each node 1 at a time. Start with 203, then 204 and then 205. Remember if all three nodes go down you will have to bootstrap one of those nodes before you can start the other. Let me know what happens.

        Comment

        Working...
        X