The class initialization of NodeId triggers a read to a system table. When using the class introduced in
CASSANDRA-2911, this happens to be triggered even though this is not necessary. This a a pain because when you use those classes, you don't have to have to set reasonable values for the data_dir and commitlog_dir in cassandra.yaml (it is a pain to have to have a cassandra.yaml in the classpath in the first place, but that's for another ticket).
This ticket proposes using lazy initialization in NodeId to avoid that.
|Transition||Time In Source Status||Execution Times||Last Executer||Last Execution Date|
|27s||1||Sylvain Lebresne||27/Jul/11 09:59|
|6h 22m||1||Sylvain Lebresne||27/Jul/11 16:21|
|Workflow||patch-available, re-open possible [ 12751725 ]||reopen-resolved, no closed status, patch-avail, testing [ 12757883 ]|
|Workflow||no-reopen-closed, patch-avail [ 12623705 ]||patch-available, re-open possible [ 12751725 ]|
|Status||Patch Available [ 10002 ]||Resolved [ 5 ]|
|Resolution||Fixed [ 1 ]|
|Status||Open [ 1 ]||Patch Available [ 10002 ]|