[teiid-issues] [JBoss JIRA] (TEIID-3786) Cache hint TTL invalidates cache before the reload/refresh is complete

[Steven Hawkins (JIRA)]

    [ https://issues.jboss.org/browse/TEIID-3786?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13121678#comment-13121678 ] 

Steven Hawkins commented on TEIID-3786:
---------------------------------------

I'm not sure I follow this issue.

> From the customer logs, you can see that the Matview is invalidated before it reloads[1]

The first line indicates that the is an invalidating refresh.  Lines 2-5 are an expected progression to loading.

Then there is an error, which presumably aborts the refresh.

Is this expected to not be invaliding?  Or is there something unexpected happening after the failed refresh?

> Cache hint TTL invalidates cache before the reload/refresh is complete
> ----------------------------------------------------------------------
>
>                 Key: TEIID-3786
>                 URL: https://issues.jboss.org/browse/TEIID-3786
>             Project: Teiid
>          Issue Type: Bug
>          Components: Server
>    Affects Versions: 8.7.1.6_2
>            Reporter: Debbie Steigner
>            Assignee: Steven Hawkins
>
> Cache hint TTL invalidates cache before the reload/refresh is complete, if the reload fails so do the user queries because the cache is invalid.
> According to the Caching documentation when using Cache Hint:
> Important
> In this example, the user query started a new refresh/load job, however it did not invalidate the current results. So, in the mean time while the load/refresh job is running, the user queries continue to return the expired results. When the refresh/reload job completes, then the old results will be flushed and new results will be used for any future user requests.
> The main idea here is not to block user queries while the refresh job is running. This is mainly due to the fact that you can not predict how long a refresh job going to run. 
> From the customer logs, you can see that the Matview is invalidated before it reloads[1], so if the reload fails for some reason then the user queries fail.  I assume this is a bug not a documentation error
> [1]
> 11:16:08,073 DEBUG [org.teiid.PROCESSOR.MATVIEWS] (Worker2_QueryProcessorQueue8) org.teiid.query.tempdata.GlobalTableStoreImpl$MatTableInfo at 3be0afa0 invalidating
> 11:16:08,074 DEBUG [org.teiid.PROCESSOR.MATVIEWS] (Worker2_QueryProcessorQueue8) org.teiid.query.tempdata.GlobalTableStoreImpl$MatTableInfo at 3be0afa0 setting matState to NEEDS_LOADING null 1445613368074 old values LOADED false
> 11:16:08,074 DEBUG [org.teiid.PROCESSOR.MATVIEWS] (Worker2_QueryProcessorQueue8) org.teiid.query.tempdata.GlobalTableStoreImpl$MatTableInfo at 3be0afa0 invalidating
> 11:16:08,074 DEBUG [org.teiid.PROCESSOR.MATVIEWS] (Worker2_QueryProcessorQueue8) org.teiid.query.tempdata.GlobalTableStoreImpl$MatTableInfo at 3be0afa0 setting matState to LOADING null 1445613368074 old values NEEDS_LOADING false
> 11:16:08,074 INFO  [org.teiid.PROCESSOR.MATVIEWS] (Worker2_QueryProcessorQueue8) TEIID30013 Loading materialized view table #MAT_MBDVCONF.CLIENT
> 11:16:08,076 DEBUG [org.teiid.CONNECTOR] (Worker3_QueryProcessorQueue9) /RMoMwxJGpmf.1.12.3 Processing NEW request: SELECT g_0.ID, g_0.NAME, g_0.CREATE_DT FROM mbdvconf_sm.CLIENT AS g_0
> 11:16:26,027 WARN  [org.jboss.jca.adapters.jdbc.local.LocalManagedConnectionFactory] (Worker3_QueryProcessorQueue9) Destroying connection that is not valid, due to the following exception: oracle.jdbc.driver.T4CConnection at 6d07b3ed: java.sql.SQLException: pingDatabase failed status=-1
> 	at org.jboss.jca.adapters.jdbc.extensions.oracle.OracleValidConnectionChecker.isValidConnection(OracleValidConnectionChecker.java:68)
> 	at org.jboss.jca.adapters.jdbc.BaseWrapperManagedConnectionFactory.isValidConnection(BaseWrapperManagedConnectionFactory.java:1100)



--
This message was sent by Atlassian JIRA
(v6.4.11#64026)