2011-09-13

• <外國講者> ​http://es.gnu.org/~reneme/fsmap/

Open data

• ​App Star 高手爭霸戰
• ​臺北市政府 vs Data.Taipei
• ​http://taipeiogdi.cloudapp.net/DataCatalog/DataSetList - 台北市公開政府資料平臺-開發人員專區資料服務(OGDI)
  • cloudapp.net 這個網域是 Microsoft Azure 平台在用的～看樣子台北市政府的開放平台是改自 ​http://ogdisdk.cloudapp.net/

Hadoop & GlusterFS

• ​Gluster Goes After Hadoop Big Data - GlusterFS 也將變成 Hadoop 支援的底層檔案系統！酷～

  the final release of GlusterFS 3.3 is expected before the end of year 2011. The new release 
  provides an integration point for Apache Hadoop enabling Hadoop users to use Gluster for storage.
  

• ​https://github.com/gluster/hadoop-glusterfs - 原始碼

OpenNebula

• First, setup ONE_AUTH

    root@controller> su - oneadmin
    oneadmin@controller> vi $HOME/.one/one_auth
  -------------------
  oneadmin:<password>
  -------------------
  

  • start opennebula service

      root@controller> /etc/init.d/opennebula start
    

• Add a new host (node) in OpenNebula pool
  1. Install 'opennebula-node' debian package in destination host

       root@node01> apt-get install opennebula-node
     

  2. Copy controller ssh key to node01

       root@controller> su - oneadmin
       oneadmin@controller> cat $HOME/.ssh/id_rsa.pub
       <COPY YOUR CONTROLLER SSH PUBKEY>
     
       root@node01> su - oneadmin
       oneadmin@node01> vi $HOME/.ssh/authorized_keys
       <PAST YOUR CONTROLLER SSH PUBKEY>
     

  3. Test an connection from controller to node01

       root@controller> su - oneadmin
       oneadmin@controller> ssh oneadmin@node01
         The authenticity of host 'node01 (192.168.0.2)' can't be established. 
         RSA key fingerprint is [].
         Are you sure you want to continue connecting (yes/no)? [yes]
         Warning: Permanently added 'node01,192.168.0.2' (RSA) to the list of known hosts.
       oneadmin@node01>
     

  4. Add this host into OpenNebula pool

       oneadmin@controller> onehost add node01 im_kvm vmm_kvm tm_ssh
     

  5. Check node registration on controller

       oneadmin@controller> onehost list
       ID NAME              CLUSTER  RVM   TCPU   FCPU   ACPU    TMEM    FMEM STAT
        0 node01            default    0    400    388    400    3.9G    2.5G   on