Local Gateway Bug: Full recovery when recovering more than one index within a cluster might fail

Hi,

Just discovered and fixed a major bug with local gateway:
http://github.com/elasticsearch/elasticsearch/issues/issue/409. A fix for it
is already in master.

-shay.banon

We've been experimenting several failures, maybe due to this error.

I've two complete work directories that flunked, in case you want to analyse
something.

When are are you planning to release 0.12 ? :slight_smile:

On Tue, Oct 5, 2010 at 7:59 PM, Shay Banon shay.banon@elasticsearch.comwrote:

Hi,

Just discovered and fixed a major bug with local gateway:
Local Gateway: Possible failure to allocate shards to nodes when more than one index exists in the cluster (on full cluster restart) · Issue #409 · elastic/elasticsearch · GitHub. A fix for
it is already in master.

-shay.banon

On Wed, 2010-10-06 at 00:59 +0200, Shay Banon wrote:

Hi,

Just discovered and fixed a major bug with local gateway:
Local Gateway: Possible failure to allocate shards to nodes when more than one index exists in the cluster (on full cluster restart) · Issue #409 · elastic/elasticsearch · GitHub. A fix
for it is already in master.

Does this warrant a 0.11.1 release before 0.12.0?

clint

Good questions, I am still not sure. Its either going to mean a 0.11.1
release next week, or bump 0.12 by a week or two. I prefer not to release a
version today, since I want to let it iron out a bit more.

-shay.banon

On Wed, Oct 6, 2010 at 2:26 PM, Clinton Gormley clinton@iannounce.co.ukwrote:

On Wed, 2010-10-06 at 00:59 +0200, Shay Banon wrote:

Hi,

Just discovered and fixed a major bug with local gateway:
Local Gateway: Possible failure to allocate shards to nodes when more than one index exists in the cluster (on full cluster restart) · Issue #409 · elastic/elasticsearch · GitHub. A fix
for it is already in master.

Does this warrant a 0.11.1 release before 0.12.0?

clint