hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jean-Daniel Cryans <jdcry...@apache.org>
Subject Re: flushing memstore when cluster shuts down WAS: empty table after flush
Date Fri, 24 Sep 2010 23:54:08 GMT
Like that:
http://hbase.apache.org/docs/r0.89.20100726/xref/org/apache/hadoop/hbase/regionserver/HRegion.html#519

(this is a link to the 0.89 code, but 0.20 works exactly the same way)

J-D

On Fri, Sep 24, 2010 at 4:49 PM, Ted Yu <yuzhihong@gmail.com> wrote:
> When HBase cluster shuts down, how is memstore flushing triggered ?
>
> Thanks
>
> On Fri, Sep 24, 2010 at 4:01 PM, Jean-Daniel Cryans <jdcryans@apache.org>wrote:
>
>> > StripedHBaseTable is just a wrapper around HTable.
>>
>> Ok.
>>
>> > QA installed another application which queries HBase. When QA didn't see
>> > meaningful data from the application.
>> > Why didn't HBase serve data from Memstore ?
>>
>> I cannot answer this question, as far as I know 0.20.6 doesn't have a
>> bug that prevents you from writing data (and many people use it in
>> production environments). My guess is, the data never got into HBase
>> or it was written with wrong timestamps and is hidden by delete
>> markers or something "time-travelly" like that. If they do play with
>> timestamps, ask them not to or verify they do the right thing.
>>
>> > QA used stop-hbase.sh to stop HBase. Would Memstore contents be flushed
>> to
>> > Storefile at that moment ?
>>
>> Yep, if you don't see any store file then the data never made it into
>> hbase.
>>
>> >
>> > QA servers were running JVM version 1.6.0_18
>>
>> From
>> http://hbase.apache.org/docs/r0.20.6/api/overview-summary.html#requirements
>>
>> "Java 1.6.x, preferably from Sun. Use the latest version available
>> except u18 (u19 is fine)."
>>
>> J-D
>>
>

Mime
View raw message