Dick Davies wrote:
I'm building a LAMP stack with Puppet, so far it's been great for
doing the 'roads and sewers' standard build bits.
I'm a bit unsure how to use it to manage e.g. the MySQL active-passive
pair we plan to use, though. Need a bit of guidance from veterans I think.
It's not HAed, we just want a spare DB node I can point the php apps at
while patching the primary.
If I was to apply a 'master' class to one node and a 'slave' to the other,
how would failover be managed?
Is that something outside Puppets reach, or would you switch the roles
in manifests, or turn off Puppet while things fail over, or something else?
There's a similar issue with NFS docroots; I have 2 NFS backend servers
that I can see from my webservers, and it's easy to tell puppet to manage the
mount. But I'm not sure whether I should be switching the mounts via Puppet
manifests, or another mechanism.
I suggest you look for another solution; Puppet best manages resources
that are *idempotent*. Not to mention the fact that you really need a
process which provides active feedback.
As an aside, for MySQL manual failover I suggest you use master-master
with a heartbeat and VIP (with a circuit breaker for managing the db
pool). Otherwise you're going to have to perform a restart of your web
cluster every time you do the flip. And unless your site is completely
read-only, you'll be stuck with data inconsistencies.
-scott
--
You received this message because you are subscribed to the Google Groups "Puppet
Users" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.
