GitHub user aledsage opened a pull request:
https://github.com/apache/brooklyn-server/pull/902
BROOKLYN-545: Fix DynamicCluster.zoneFailureDetector default val
Fixes https://issues.apache.org/jira/browse/BROOKLYN-545
Below is an extract from the comments in the code, explaining it - see
other code comments for more info:
```
// Previously, when there was a mutable default value for
ZONE_FAILURE_DETECTOR, the second cluster
// would include failure info for the first cluster's zone-locations.
When these were unmanaged,
// the zone-locations would become dangling referenes within the
ZoneFailureDetector. This caused
// NPE on rebind.
//
// Config key persistence was then changed so that statically defined
keys (such as ZONE_FAILURE_DETECTOR,
// declared on the DynamicCluster class) would not be persisted. This
meant that the ZoneFailureDetector
// state was no longer persisted for either location. The NPE on rebind
went away, but the config key
// default value instance was still shared between all clusters.
//
// Now the detector has a null default value. We implicitly set a
transient field for the zoneFailureDetector
// so if using defaults then nothing is persisted. But if a custom
ZoneFailureDetector is supplied, then
// that will be persisted and used.
```
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/aledsage/brooklyn-server
fix-DynamicCluster-zoneFailureDetector-default
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/brooklyn-server/pull/902.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #902
----
commit 76ab91489d04cac693399d1b9a1ba1bbb53fa1ae
Author: Aled Sage <[email protected]>
Date: 2017-10-20T16:29:06Z
BROOKLYN-545: Fix DynamicCluster.zoneFailureDetector default val
----
---
