I have tried to set replication factor to 1 by hadoop fs -setrep -R 1
/ but no luck
On Fri, Oct 31, 2014 at 1:03 PM, Li Li wrote:
> my hbase(phoenix) crashes when the disk is full. I killed the hbase
> and hadoop related processes. But After that I can't start hbase
> anymore.
> I am using ubuntu
my hbase(phoenix) crashes when the disk is full. I killed the hbase
and hadoop related processes. But After that I can't start hbase
anymore.
I am using ubuntu 12.04 and hadoop-1.2.1 and hbase 0.98.5 with phoenix 4.1.0
The region server print error like:
2014-10-31 12:33:49,835 INFO
[mobvoi-knowle
Hi Mike,
Yes, the bit configuration is different in Phoenix to ensure that the
sort order of the binary representation matches the natural sort order
of the data. This improves Phoenix performance by allowing many more
query optimizations than would otherwise be possible.
You can convert back/fort
Hi,
It seems like the Phoenix formats for numerics like integer, float, are
different, in terms of the bit configurations, than the way non-Phoenix HBase
clients store and interpret numerics. Unsigned numbers are the same, but when
they’re signed the representations appear to be different. Is t
i want to know whether can phoenix do this:
for two condition in where clause;
first filter the condition1 by index on col1 and get the result set;
Then filter the result set with condition2 by index on col2;
and get the total result ultimately。
if can do this ,phoenix will prefect in such sit
Phoenix client set some special attributes like index details to mutations
based on that index mutations will be prepared to add to index table.
But when you use the tool that might be missing I think that's why index tables
might not have any data.
Is it not possible to directly load into Phoen
another problem is :
1.i create table use phoenix client,
2.load data to Hbase by spoon kettle ,
3,i can get data from phoenix client,
4 then i create index on cols,
5 load another group data to Hbase by spoon;
6 i can't get the new data in step 5;
whether only use phoenix client to insert data c
Currently Phoenix doesn't have support of index merge or kind of that.
I think you have created separate index on each column and your query has where
clause with conditions on multiple columns.
if we take any index which will not have all the columns in the where clause.
That's why no index is u
the where conditions are all have index,e.g (col2=1 and col3=1 . col1 to coln
all have local indexes),
when :
select col2 from table where col2=1 and col3=1, it will full scan the table;
only if one condition after "where" clause, it will scan the index_table;
-- Original
If you have where condition on non indexed column then we go for full table
scan.
Can you confirm this?
Thanks.
Rajeshbabu.
This e-mail and its attachments contain confidential information from HUAWEI,
which
is intended only for the person or entity whose addre
table:
each col with local index(col1..coln),
when i select data such as :
select col1,col2 from table where col1="test", the index works,
but if there are 2 or more conditions after "where" clause,the query will full
scan the table;
is there any way to use index for such query?
thanks.
xux
11 matches
Mail list logo
