Hi,
we use user_id to join the fact table and join table,so the user_id is set to
dimension column, but we don't want to aggregate data by user_id, it's a high
cardinality column,can we avoid it by just make one aggregation groups without
user_id, will kylin ignore user_id in that case?
we have some problems in hadoop environment:
1、how does kylin find the hadoop environment ? if we export such as
HIVE_HOME=/root/hive_2.10-0.10.0.0 into /etc/profile,could the profile file
help kylin to find hive or hbase environment?
2、we have already installed spark2 in the cluster ,how to use
se the standard Date
>format in Hive. Or you can implement your own logic, with the interface "
>IPartitionConditionBuilder"
>
>2017-10-10 11:33 GMT+08:00 崔苗 :
>
>> well,the timestamp column was bigint such as 1507547479434 in hive table,
>> when I define the e
he cube is partitioned
>by day/week/month; in some cases, it is by the hour; In streaming case, it
>might partition by the minute; But no case by timestamp. I put some
>comments about this in this document:
>https://kylin.apache.org/docs21/tutorial/cube_streaming.html
>
>2017-10-0
so what about hive,hbase and spark ? just looks for library too ?
在 2017-10-09 18:28:08，Billy Liu 写道：
>Kylin looks for Kafka library from the classpath, not the Kafka broker
>servers. You could deploy kafka into any other servers.
>
>2017-10-09 16:30 GMT+08:00 崔苗 :
>
>> Hi,
Hi,
we have some problems in kylin installation , we want to know how does kylin
find the hbase, hive or kafka environment , how to configure kylin if these
servers such as kafka are installed in other cluster, not the same machine with
kylin.
thanks in advanced for your reply.
Hi,
we want to use tables in kafka as fact tables and tables in MySql as lookup
tables,so we put all the tables into hive and want to join them as cubes.
the time column in fact table was timestamp, so does kylin2.1 support timestamp
for cube partition?
I find this :https://issues.apache.org/jir
the timestamp in kafka table can be recognized and automatically be changed
into date format,but the timestamp in hive can't be used?
在 2017-09-30 16:24:43，"Alberto Ramón" 写道：
>I think yes. You have 3 format to choose o column sompose
>
>On 30 September 2017 at 09:21,
mm:ss
>
>Check this JIRAS:
>https://issues.apache.org/jira/browse/KYLIN-1101
>https://issues.apache.org/jira/browse/KYLIN-1441
>https://issues.apache.org/jira/browse/KYLIN-1427
>
>On 30 September 2017 at 04:45, 崔苗 wrote:
>
>> we have a timestamp column used as the
we have a timestamp column used as the partition column,what should be used to
encoed the column,date or time? what's the difference between the two code?
BTW, can kylin recognize all the level of timestamp ,no matter it's s or ms
timestamp?
thanks in advanced for your reply.
how about int column? Can I use a integer column in on cube both as dimension
and distinct count?
在 2017-09-29 11:41:02，ShaoFeng Shi 写道：
>Split that into two cubes: in one it acts as a dimension, in the other, it
>acts as count distinct measure.
>
>2017-09-29 11:12 GMT+08:00 崔苗
if a column was used both as dimension and distinct count,what should be use
to encode ?
在 2017-09-29 11:05:28，ShaoFeng Shi 写道：
>Use GD for dimension encoding will cause query inaccurate, that should be
>avoided.
>
>Please use dict or fixed_length encoding for that column.
>
>2017-09-29 10:37 GM
ith path conf/.keystore due to
> /root/apache-kylin-2.1.0-bin-hbase1x/tomcat/conf/.keystore (没有那个文件或目录)
Just like the error says, there is a missing config file at
KYLIN_HOME/tomcat/conf/.keystore
On Tue, Sep 19, 2017 at 4:18 PM, 崔苗 wrote:
Hi,all :there were some errors in starting kylin instance
Hi,all :there were some errors in starting kylin instance.hadoop
version:2.7.3.2.6.2.0-205 installed by ambarikylin version :2.1.0two kylin
instances:one:all mode ,the other : query modekyin instance started but can't
open webui,no service in 7070 port
kylin.out:九月 19, 2017 3:21:14 下午
org.apach
we have some problems in using kylin api:1 the wipe cache doesn't work,we want
to 'drop' the 'kylin_sales_cube',the response status_code was 200,but the cube
is still ready2 we want to build job by hour, and there were a dozen of
cubes,so the jobs were building parallelly,how to manage these job
Hello Folks:
we failed to build streaming cube in first step:save data from kafka, this is
the output log:Counters: 12
Job Counters
Failed map tasks=4
Launched map tasks=4
Other local map tasks=3
Rack-local map tasks=1
16 matches
Mail list logo
