[
https://jira.jboss.org/jira/browse/ISPN-120?page=com.atlassian.jira.plugi...
]
Galder Zamarreno commented on ISPN-120:
---------------------------------------
As per conversation on IRC:
- L1 entries should not go into the cache store. Hence, cache store interceptor should
detect if the put is a L1 put. Might required subclassing the cache store for DIST mode.
- Normally, when replication and shared cache stores are in use, the cache instance where
the data is put is responsible for storing it in the cache store. This is not necessary
the case for a distributed cache. In such scenarios, the owner of the key should be
responsible to store it.
Inmortal cache entries stored as mortal entries in cache store
--------------------------------------------------------------
Key: ISPN-120
URL:
https://jira.jboss.org/jira/browse/ISPN-120
Project: Infinispan
Issue Type: Bug
Components: Loaders and Stores
Affects Versions: 4.0.0.ALPHA5
Reporter: Galder Zamarreno
Assignee: Galder Zamarreno
Fix For: 4.0.0.BETA1
Let's say you have a cluster of 3 nodes: n1, n2, and n3, you're using a shared
cache sotre and and you store 1=one inmortal k,v pair in n1,
Assume that the owners should be n2 and n3. There appears to be a bug in which n2 stores
the inmortal entry as mortal with default lifespan for L1.
This arose during demo testing. Needs further investigation.
Another thing to check is purging of expired entries in cache store. When an expired
entry is loaded, it's not purged from the store, maybe it should do so??
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see:
http://www.atlassian.com/software/jira