Author: scottmf
Date: 2009-11-12 10:14:37 -0800 (Thu, 12 Nov 2009)
New Revision: 13939
URL: http://svn.hyperic.org/?view=rev&root=Hyperic+HQ&revision=13939
Modified:
trunk/src/org/hyperic/hq/measurement/server/session/AvailabilityManagerEJBImpl.java
Log:
[HHQ-3505] fix potential for NonUniqueObjectExceptions in addData() flow
Modified: trunk/src/org/hyperic/hq/measurement/server/session/AvailabilityManagerEJBImpl.java
===================================================================
--- trunk/src/org/hyperic/hq/measurement/server/session/AvailabilityManagerEJBImpl.java 2009-11-11 20:54:40 UTC (rev 13938)
+++ trunk/src/org/hyperic/hq/measurement/server/session/AvailabilityManagerEJBImpl.java 2009-11-12 18:14:37 UTC (rev 13939)
@@ -779,10 +779,18 @@
for (Iterator it=_removeMap.entrySet().iterator(); it.hasNext(); ) {
Map.Entry entry = (Map.Entry)it.next();
AvailabilityDataRLE rle = (AvailabilityDataRLE)entry.getValue();
+ // if we call remove() on an object which is already in the session
+ // hibernate will throw NonUniqueObjectExceptions
+ if (_dao.getSession().contains(rle)) {
+ _dao.getSession().evict(rle);
+ }
_dao.remove(rle);
}
- // for some reason if flush is not run, then create() will throw
- // Hibernate NonUniqueObjectExceptions
+ // addData() could be overwriting RLE data points (i.e. from 0.0 to 1.0)
+ // with the same ID. If this is the scenario, then we must run
+ // flush() in order to ensure that these old objects are not in the
+ // session when the equivalent create() on the same ID is run,
+ // thus avoiding NonUniqueObjectExceptions
_dao.getSession().flush();
for (Iterator it=_createMap.entrySet().iterator(); it.hasNext(); ) {
Map.Entry entry = (Map.Entry)it.next();
|
