----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://review.hbase.org/r/211/#review260 -----------------------------------------------------------
Ship it! I already +1'd this patch over in the issue (this review.hbase.org seemed down for me this morning when I went to review). - stack On 2010-06-20 23:11:25, Todd Lipcon wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://review.hbase.org/r/211/ > ----------------------------------------------------------- > > (Updated 2010-06-20 23:11:25) > > > Review request for hbase, stack and Mingjie Lai. > > > Summary > ------- > > The issue was that when specifying an explicit start row of EMPTY_BYTE_ARRAY > for a metascan with a specified table, it would try to call > getClosestRowBefore with the region ID set to ZEROES instead of NINES. So, > the first region would be invisible to the getClosestRowBefore call, and it > would incorrectly decide the table did not exist. > > > This addresses bug HBASE-2760. > http://issues.apache.org/jira/browse/HBASE-2760 > > > Diffs > ----- > > src/main/java/org/apache/hadoop/hbase/client/MetaScanner.java 7fe6f63 > src/test/java/org/apache/hadoop/hbase/client/TestMetaScanner.java > PRE-CREATION > > Diff: http://review.hbase.org/r/211/diff > > > Testing > ------- > > New unit test fails without fix. Also verified that the warnings are gone > with my client program. > > > Thanks, > > Todd > >
