-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/6114/#review9385
-----------------------------------------------------------


 I am wondering if there is a way to invalidate a cache entry and instantiate a 
new one whenever there is a connection failure. It will not be very easy. We 
can probably reduce the default cache time to 2-3 minutes for now and look at 
advanced cache invalidation later. 


http://svn.apache.org/repos/asf/incubator/hcatalog/branches/branch-0.4/src/java/org/apache/hcatalog/common/HCatUtil.java
<https://reviews.apache.org/r/6114/#comment20119>

    The HiveClientCache can be a singleton. The factory leaves the possibility 
of multiple caches being initialized.



http://svn.apache.org/repos/asf/incubator/hcatalog/branches/branch-0.4/src/java/org/apache/hcatalog/common/HCatUtil.java
<https://reviews.apache.org/r/6114/#comment20117>

     Create the MetaException with the full exception instead of 
e.getMessage(). Will be difficult to determine the actual reason with just the 
message.
    
    spelling mistake - coulnd't.



http://svn.apache.org/repos/asf/incubator/hcatalog/branches/branch-0.4/src/java/org/apache/hcatalog/common/cache/HiveClientCache.java
<https://reviews.apache.org/r/6114/#comment20120>

    Same here too. Would be better to use the base exception in constructor and 
not just the message.



http://svn.apache.org/repos/asf/incubator/hcatalog/branches/branch-0.4/src/java/org/apache/hcatalog/common/cache/HiveClientCache.java
<https://reviews.apache.org/r/6114/#comment20121>

    Can we make this a Throwable so that we can ensure there are no errors in 
finalize.


- Rohini Palaniswamy


On July 23, 2012, 11:28 p.m., Arup Malakar wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/6114/
> -----------------------------------------------------------
> 
> (Updated July 23, 2012, 11:28 p.m.)
> 
> 
> Review request for hcatalog and Rohini Palaniswamy.
> 
> 
> Description
> -------
> 
> Time expiry based HiveMetaStoreClient cache for HCatalog
> 
> 
> This addresses bug HCATALOG-370.
>     https://issues.apache.org/jira/browse/HCATALOG-370
> 
> 
> Diffs
> -----
> 
>   
> http://svn.apache.org/repos/asf/incubator/hcatalog/branches/branch-0.4/ivy.xml
>  1364830 
>   
> http://svn.apache.org/repos/asf/incubator/hcatalog/branches/branch-0.4/src/java/org/apache/hcatalog/common/HCatUtil.java
>  1364830 
>   
> http://svn.apache.org/repos/asf/incubator/hcatalog/branches/branch-0.4/src/java/org/apache/hcatalog/common/cache/CacheFactory.java
>  PRE-CREATION 
>   
> http://svn.apache.org/repos/asf/incubator/hcatalog/branches/branch-0.4/src/java/org/apache/hcatalog/common/cache/HiveClientCache.java
>  PRE-CREATION 
>   
> http://svn.apache.org/repos/asf/incubator/hcatalog/branches/branch-0.4/src/java/org/apache/hcatalog/mapreduce/DefaultOutputCommitterContainer.java
>  1364830 
>   
> http://svn.apache.org/repos/asf/incubator/hcatalog/branches/branch-0.4/src/java/org/apache/hcatalog/mapreduce/FileOutputCommitterContainer.java
>  1364830 
>   
> http://svn.apache.org/repos/asf/incubator/hcatalog/branches/branch-0.4/src/java/org/apache/hcatalog/mapreduce/FileOutputFormatContainer.java
>  1364830 
>   
> http://svn.apache.org/repos/asf/incubator/hcatalog/branches/branch-0.4/src/java/org/apache/hcatalog/mapreduce/HCatOutputFormat.java
>  1364830 
>   
> http://svn.apache.org/repos/asf/incubator/hcatalog/branches/branch-0.4/src/java/org/apache/hcatalog/mapreduce/InitializeInput.java
>  1364830 
>   
> http://svn.apache.org/repos/asf/incubator/hcatalog/branches/branch-0.4/src/java/org/apache/hcatalog/pig/PigHCatUtil.java
>  1364830 
>   
> http://svn.apache.org/repos/asf/incubator/hcatalog/branches/branch-0.4/src/test/org/apache/hcatalog/common/TestHiveClientCache.java
>  PRE-CREATION 
> 
> Diff: https://reviews.apache.org/r/6114/diff/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Arup Malakar
> 
>

Reply via email to