Find the answer to your Linux question:
Results 1 to 4 of 4
Hello everybody, I am going to setup a failover solution, that is 2 db server (InnoDB as storage engine)with the second starting when the first crashes, using Red Hat Cluster ...
Enjoy an ad free experience by logging in. Not a member yet? Register.
  1. #1
    Just Joined!
    Join Date
    Mar 2006
    Posts
    16

    Smile MySQL InnoDB Failover and Red Hat Cluster Suite


    Hello everybody,
    I am going to setup a failover solution, that is 2 db
    server (InnoDB as storage engine)with the second
    starting when the first crashes, using Red Hat Cluster
    Suite and a SAN to store the data files.
    I would like to know if I must enable external-locking
    option on my.cnf file. I am doubtful about it, maybe
    it is not necessary, Red Hat Cluster Suite should
    handle everything and InnoDB locks anyway the data
    files, right?
    Thanks for any help.

    Stefano

  2. #2
    Just Joined!
    Join Date
    Oct 2006
    Posts
    2
    if only one mysql serveur is up at a time you don't need to set the external locking because noone else is writing on the shared file system !

  3. #3
    Just Joined!
    Join Date
    Mar 2006
    Posts
    16

    Smile InnoDB help

    I think so, but better safe than sorry, this is the reason I asked for. As a matter of fact I have not found documentation about setting up a failover solution with Red Hat Cluster Suite and InnoDB (I have only found how to set up MySQL services by Red Hat Cluster Suite, but using InnoDB tables is quite different that using MyISAM Tables: InnoDB engine is transaction-safe and has an autorecovery mechanism) I wonder if you or someone else in this forum could give any help (links, whitepapers, any kind of advise would be really precious!!!!).
    Have a nice day

  4. #4
    Just Joined!
    Join Date
    Mar 2006
    Posts
    16

    Cool Ok!!

    OK, it works properly, at least it seems so!
    For this forum's friends, a brief summary.
    Do not use external-locking, only a server writes.
    Binlog file is in the shared storage, too.
    In case of crash of the active node, there is a switch to the stand-by node. The autorecovery starts, rollbacks the uncommitted transactions and InnoDB is started too.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •