On Aug 3, 9:05am, Phil Stracchino wrote:
} On 08/03/18 04:01, John Nemeth wrote:
} > On Jul 26, 8:29am, Phil Stracchino wrote:
} > } Are you talking about Oracle's MySQL Cluster that has gone through half
} > } a dozen names in the last ten years and was formerly known as NDB
} > } Cluster?
} >
} > I am talking about MySQL Cluster.
} >
} > I first started investigating it about four years ago when I
} > was looking for a redundant database solution for a project. I
} > haven't seen any name changes in that time, but I can't speak to
} > what may have happened prior. The engine it uses is called NDBCLUSTER
} > (or NDB for short).
}
} OK, we're on the same page then. Oracle has been renaming it every few
} years. As you've noted, the engine behaves *similarly* but not
} identically to InnoDB. I imagine it could be made to work well, but
} work would definitely be required.
}
} Getting Bacula to work with a Galera cluster on the other hand is pretty
} straightforward. At the moment the principal thing you need to do for
} that is turn off attribute spooling, because the writeset limit in the
} attribute spooling code doesn't work.
Ah, this would be the problem I ran across. I made the changes
to the schema and tried some test runs. That was okay, but when
I tried it with production backups it would fail when despooling
attributes. My C++ is pretty much non-existant (I spend my time
digging around inside kernels) so I ended up giving up and installing
yet another instant of mysqld on the box running the director.
}-- End of excerpt from Phil Stracchino
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Bacula-devel mailing list
Bacula-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-devel
