[
https://issues.jboss.org/browse/TEIID-1689?page=com.atlassian.jira.plugin...
]
Ramesh Reddy commented on TEIID-1689:
-------------------------------------
To test
1) Configure 2 nodes in a cluster, but start only single node Node1
2) Populate the Node1 by querying a materialized view
3) Start Node2, wait until it is completely started has time to sync with other node.
4) Kill/Stop Node1
5) Query Node2 for the same view that has been Materialized on Node1, and make sure that
the query did not query the source to deliver the results. A good way to do this is to
provide a bad password in the -ds.xml file for the data source in Node2
6) If you see results, this is working, otherwise log JIRA.
Load internal mateirlized view cached contents when node joins the
cluster
--------------------------------------------------------------------------
Key: TEIID-1689
URL:
https://issues.jboss.org/browse/TEIID-1689
Project: Teiid
Issue Type: Feature Request
Components: Server
Affects Versions: 7.4.1
Reporter: Ramesh Reddy
Assignee: Ramesh Reddy
Fix For: 7.4.1, 7.5
When IMV loaded at Node 1, and the contents of the that IMV will be loaded at Node 2 only
when the query is issued against it. Then user incurs the wait time to load the contents
from Node 1. If in the situation if Node 1, went down before the query to Node 2, in that
case node 2 has to fetch from the source and cached contents will be lost.
This can be avoided by hot loading the all the cached contents at the start of the node
from other nodes.
--
This message is automatically generated by JIRA.
For more information on JIRA, see:
http://www.atlassian.com/software/jira