On Jan 2, 2008 11:12 AM, Luis Motta Campos <luismottacampos@stripped> wrote:
> Baron Schwartz wrote:
> > On Jan 2, 2008 10:04 AM, Luis Motta Campos wrote:
> >> Baron Schwartz wrote:
> >>> What are the biggest changes you anticipate? I'd say they will be
> >>> the version upgrade, converting to InnoDB, and using
> >>> replication. It looks like you have planned well for all but
> >>> using replication.
> >> That's interesting. What kind of activities (besides configuration,
> >> maybe?) are interesting to add?
> > Lots of experimentation!
> > Let me ask it another way: how are you planning to use replication?
> > -- for load balancing/scaling reads, backups, a hot standby machine?
> OK, I guess a bit more of information can help. I have a very sensitive
> database at the company
> The master/spare and slave/spare machines would be connected through
> heartbeat and will keep the database in a DRDB filesystem. This will
> guarantee that, in case of failure, the other machine can raise the same
> IP address in it's own interface and continue operations after a short
I suspect that this won't go smoothly on the first try, and if it's
very sensitive... No offense, but you're talking about a lot of
complexity here, and you're not familiar with replication yet. DRBD,
failover etc are even more complex.
I personally would hire an expert to help me plan and implement this
upgrade. MySQL AB has consultants, but there's also Pythian, Proven
Scaling, and Percona, to name a few whom I think are very familiar
with the tools and technologies you're talking about.
(Why do the consulting companies all have names that start with P, I wonder?)