-
Type:
Bug
-
Status: Closed
-
Resolution: Fixed
-
Affects Version/s: Master
-
Fix Version/s: 7.0.0 DXP FP28, 7.0.X EE, 7.0.4 CE GA5, 7.1.X, Master
-
Component/s: Performance > Caching Infrastructure
-
Branch Version/s:7.0.x
-
Backported to Branch:Committed
-
Story Points:1.75
-
Git Pull Request:
Steps to reproduce:
1) Set up cluster of two nodes
2) Enable bootstraploader on two nodes by setting:
ehcache.bootstrap.cache.loader.enabled=true
3) Enable to load caches in a sync way on two nodes by setting:
ehcache.bootstrap.cache.loader.properties.default=bootstrapAsynchronously=false
4) Start node one and wait for it is fully started
Before fix, node one can not be started fully, it will be blocked.
After fix, node on can be started fully
5) Try to get a cache from MultiVMPoolUtil and put some data into it. This can be done by a jsp script on webapps/ROOT or run scripts on Control Panel > Server Administration > Script
PortalCache portalCache = MultiVMPoolUtil.getCache("test.cache");
portalCache.put("testKey1", "Test Value1!!!!");
6) Start node two and wait for it is fully started
7) Check cache entry put in node one loaded on node2, again you can check this by run scripts on Control Panel > Server Administration > Script
8) Shutdown node2
Excepted: After node2 is shutdown, check on node1, the cache value put earlier can be seen.
Actual result: After node2 is shutdown, check on node1, the cache value put earlier can not be seen.