Kresten,
Thanks for the help! I successfully joined my additional node to the
cluster and *hopefully* the rest go smoothly now.
I took your advice and stopped the node, moved the bad merge index out of
the 'merge_index' directory and then restarted the node. After watching the
log, and viewing
ommended way is to re-index all your keys to
> verify that you recreated the destroyed indexes.
>
So in order to re-index all the keys can I run the simply run the search-cmd
on the bucket w/ some params? Or do I need to use one of the clients to
literately open each key, and then re-save
Ryan Zezeski-2 wrote
>
> This is not a good msg to have. I'm not sure if it's causing you any harm
> but Riak has two notions of the ring. A per-node ring which is internal
> to
> the node and a gossiped ring which all nodes share with each other. For
> some reason, an internal ring has leaked
