Re: Unable to connect to drill 1.1.0 using JDBC

When is 1.2 planned for release? Sent from my iPhone > On Aug 18, 2015, at 2:53 PM, Paul Mogren wrote: > > I looked at JIRA and it seems this fix is sitting around waiting for a > 1.2.0 release to roll around, which looks like it is still going to be a > while longer. Could we please have a poi

Re: Unable to connect to drill 1.1.0 using JDBC

I looked at JIRA and it seems this fix is sitting around waiting for a 1.2.0 release to roll around, which looks like it is still going to be a while longer. Could we please have a point release for this? Thanks, Paul On 7/28/15, 2:42 PM, "Daniel Barclay" wrote: >Parth Chandra wrote: >> Yes a

Re: Drill Installation Question

Well, the requirement is the users will access Drill using a webui from their desktops and only few ports on edge node is open to the outside world. So, I need something running on the edge node to run drill queries. I am not a network expert but let me figure out about NAT, tunnel etc which you me

Re: Drill Installation Question

Can’t you use a network service like NAT, tunnel, etc to provide access? —Andries > On Aug 18, 2015, at 2:33 PM, Abhishek Girish > wrote: > > I don't think that is currently possible. AFAIK, the web-UI is tied to the > drillbit service and cannot be brought up stand-alone. > > On Tue, Aug 18,

Re: Drill Installation Question

I don't think that is currently possible. AFAIK, the web-UI is tied to the drillbit service and cannot be brought up stand-alone. On Tue, Aug 18, 2015 at 2:22 PM, Naveen Madhire wrote: > Thanks for all your help. One last question guys :) > > Currently, I have started drillbits on all the datano

Re: Drill Installation Question

Thanks for all your help. One last question guys :) Currently, I have started drillbits on all the datanodes and accessing the drill command line from an edge node outside of the cloudera cluster. I can see that, I can only access the webui on 8047 port from the datanodes. I need to access the w

Re: Apache drill - Cannot access HDFS data source running in another instance

I'm not sure, may be others can help. It could also be a bug. Would help if you could provide more details on your Hadoop setup (number of nodes, security, ...). I'm assuming by doing 'use dfs.tmp' followed by 'show files' on sqlline, the 'test.json' file gets listed. Also can you upload logs som

Re: Drill Installation Question

You could override defaults in drill-override.conf. Example: http: { enabled: true, port: 8047 } Refer to drill-override-example.conf for details. On Tue, Aug 18, 2015 at 1:38 PM, Naveen Madhire wrote: > Thanks for the info. > > Also, I didn't find any info on changing the default po

Re: Drill Installation Question

Thanks for the info. Also, I didn't find any info on changing the default ports 8047, 31010 etc. I need to be able to configure specific ports to be used for Drill service. Please let me know. Thanks. On Tue, Aug 18, 2015 at 3:11 PM, Andries Engelbrecht < aengelbre...@maprtech.com> wrote: > See

Re: Drill Installation Question

See https://drill.apache.org/docs/querying-system-tables/ —Andries > On Aug 18, 2015, at 11:51 AM, Naveen Madhire wrote: > > Well, this is working well now. I really like the install process, very > simple and straight forward. :) > >

Re: Drill Installation Question

Well, this is working well now. I really like the install process, very simple and straight forward. :) One have one question, I ran the select * from sys.drillbits query, I see the output gives, user port, control port, data port and current? What does these each one mean ? Thanks. On Mon,

Re: UNION infinite Pending

If the query remains infinitely in PENDING status, and you did not get a Physical Plan, that might indicate a bug in the sql planner & optimizer. You may open a JIRA, and provide your view DDL so that we can see if we could re-produce it. On Tue, Aug 18, 2015 at 10:16 AM, Boris Chmiel < boris.ch

UNION infinite Pending

I found that creating a View with more than 7 UNION clause on the same table (around 40 Operators), leads the query to remain infinitely in PENDING status on my configuration. The Physical Plan is not created. Maybe It is about parameter tuning. Any idea ?

Re: [DRILL] Column display for mongodb

Most tools today require the metadata to be available when they connect to a data source like Drill. As Drill works with multiple data sources where schema is only discovered at query time, it can create challenges for Tools that require a predefined schema at connection time. The best way to

Re: [DRILL] Column display for mongodb

Hello, Maybe the Excel client needs some metadata, and the MongoDB plugin is getting the schema on read (as you read data from your query) One solution, would be to create a view and use this view into your excel query, see https://drill.apache.org/docs/create-view/ Can you test and let us know

RE: [DRILL] Column display for mongodb

Hi Kristine, Yes, I am also able to use the example queries. I wanted to see the columns while trying from excel so that I could selectively choose some. But it is showing *. Thanks, Sudip -Original Message- From: Kristine Hahn [mailto:kh...@maprtech.com] Sent: 18 August 2015 PM 07:28

Re: [DRILL] Column display for mongodb

I can use Drill to query a Mongo table and the column output is fine as shown in http://drill.apache.org/docs/mongodb-plugin-for-apache-drill/#example-queries Why do mention MS Query? Please try the steps in http://drill.apache.org/docs/mongodb-plugin-for-apache-drill

[DRILL] Column display for mongodb

Hi, I was trying to use mongodb plugin with apache drill (1.1) using MSQuery. I could browse through the collection I have (the sample zipcode database) but I was not able get the columns displayed. What could I be missing here? Could you please help to troubleshoot? [cid:image001.png@01D0D9E5.7

Re: Tableau / filter issues

Hi usersFor people facing the same problem, here the workaround in case Tableau Quick filters does not populate. You need to enforce the column data type manually and not let Tableau do it by itself. Here is 2 approaches to do it : - create a "raw" view on top of a parquet file with typed fields

Re: Memory usage

To close this thread : After trying different OS and machines configuration, the best way to go to circumvent OOM is buying RAMBoris Le Mercredi 5 août 2015 14h01, Boris Chmiel a écrit : Hi Andries, thanks for your support Bumping the heap or disabling hash_join alone does not cau