On the same tables I get missing row keys for a mask in the prefix, if I
mask the second part of the key like this:
201401\x00\x00\x00\x00\x00_product1___
and fuzzy info:
{0,0,0,0,0,0,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0}
It seems to work
Anyone encounter issues with masking the prefix ? Seems
Looking at TestFuzzyRowFilter.java, there're several cases where mask
starts with 1.
It would be much easier to diagnose if you come up with a unit test.
Take a look at TestHRegionServerBulkLoad.java since you mentioned you're using
bulk load in the cluster.
Cheers
On Thu, Mar 13, 2014 at
I can't seem to reproduce in unit test.
The main difference is that I'm using bulk load in the cluster and Put API
in the unit test.
On Mon, Mar 10, 2014 at 4:47 PM, Ted Yu yuzhih...@gmail.com wrote:
Amit:
Can you put your scenario in a unit test so that it is easier to pinpoint
where the
Hi all,
I'm using HBase 0.94.12 + Hadoop 1.0.4.
Trying to use FuzzyRowFilter looks like it's missing keys in the scan.
Row key structure:
MMdd_Country_Product_Category1_Category2_
Where the date is mandatory and all other fields may be .
Examples:
20140101_US_product1___
Is it because you fixed _US_product1___ part of the key? From your
definition of filter you should get as output all keys of form
MMdd_US_product1___.
can you share a key thats of this format and missing in the output?
On Mon, Mar 10, 2014 at 3:38 PM, Amit Sela am...@infolinks.com wrote:
My table contains keys of this kind over an entire month but the scan
returns only for a some fo the days.
I have 2010101-20140131 but the scan returns only for:
20140104, 20140110, 20140111, 20140118, 20140120, 20140125, 20140128
Using get or scan with no fuzzy filter works...
On Mon, Mar 10,
Amit:
Can you put your scenario in a unit test so that it is easier to pinpoint
where the issue is ?
Thanks
On Mon, Mar 10, 2014 at 5:25 AM, Amit Sela am...@infolinks.com wrote:
My table contains keys of this kind over an entire month but the scan
returns only for a some fo the days.
I have
