Nick Tkach
unread,Oct 2, 2012, 3:56:51 PM10/2/12Sign in to reply to author
Sign in to forward
You do not have permission to delete messages in this group
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to mongod...@googlegroups.com
We have a fairly typical setup currently, two "groups" each set up with 3 boxes:
Prod
node1
node2
node3
Staging
nodeA
nodeB
nodeC
node1 and nodeA are 2.0 at the moment and everything else is 2.2. The
only places that there is anything running right now is a couple of
standalone mongo instances on node1 and nodeA (unrelated). They are
currently running with authentication, yes (we'd like to keep it that
way in the end result). What we want to do ultimately is get
everything over to 2.2 *and* convert the existing
databases/collections over to sharded replica sets (one cluster for
prod and one cluster for staging).
I've found documentation on doing one *or* the other, but seemingly
not both. My understanding is that we *could* stop the single node
mongods, upgrade the binary, and restart each, then go through that
process of the standalone -> cluster/shard migration, but we'd like to
avoid downtime if possible. Has anyone done this or written it up?
Or is what I'm thinking here the best/only way to do it?