Thanks Kamesh. One is sharded and the other one is not sharded. In both cases, I experienced errors when doing simple select queries (except that count was working).
In the log folder, I can only see sqlline.log and sqlline_queries.json. When I invoke drill-embedded, do I need to specify any argument to output error logs? Thank you. On Wed, Aug 26, 2015 at 4:14 PM, Kamesh <[email protected]> wrote: > Hi Xiao, > Both database and collection are sharded? If not, you get the same errors. > If both of them sharded, Could you send the error logs, if possible? > > On Wed, Aug 26, 2015 at 11:39 AM, Xiao Yang <[email protected]> wrote: > > > Thanks Kamesh. I tested it on another database that is sharded. I > > experienced errors. Are you sure that is the case? > > > > On Wed, Aug 26, 2015 at 3:44 PM, Kamesh <[email protected]> wrote: > > > > > Hi Xiao, > > > > > > There is an issue when a database or a collection is unsharded in the > > Mongo > > > DB cluster. I submitted patch also for this issue DRILL-1752 > > > <https://issues.apache.org/jira/browse/DRILL-1752> > > > > > > > > > > > > On Wed, Aug 26, 2015 at 11:09 AM, Xiao Yang <[email protected]> > > wrote: > > > > > > > Hi Kamesh, > > > > > > > > The 'zips' database is not sharded. The MongoDB setup is configured > > with > > > > two shards. There are other databases that are sharded. > > > > > > > > Thank you. > > > > Xiao > > > > > > > > On Wed, Aug 26, 2015 at 3:33 PM, Kamesh <[email protected]> > > wrote: > > > > > > > > > Hi Xiao, > > > > > Is zips collection sharded or unsharded in the mongo cluster? > > > > > > > > > > > > > > > On Wed, Aug 26, 2015 at 10:34 AM, Xiao Yang <[email protected] > > > > > > wrote: > > > > > > > > > > > Hi, > > > > > > > > > > > > I setup a test environment for using Drill with MongoDB. I had > > > problems > > > > > > getting queries to work. > > > > > > > > > > > > ======A brief description of my setup========= > > > > > > > > > > > > Mongo cluster configuration: > > > > > > 1 MongoDB router > > > > > > 1 config server > > > > > > 2 shards > > > > > > > > > > > > MongoDB versions: > > > > > > 3.0.4 > > > > > > Configured to use Wiredtiger as the storage engine > > > > > > > > > > > > The Linux version that I am using is: > > > > > > Ubuntu 14.04.2 LTS (GNU/Linux 3.13.0-54-generic x86_64 > > > > > > > > > > > > The Java version number is: > > > > > > OpenJDK 64-bit Server VM (build 24.79-b02, mixed mode) > > > > > > > > > > > > Drill version: 1.1.0 > > > > > > > > > > > > ======A brief description of the problems that I > > encountered========= > > > > > > > > > > > > 1. I imported the zips.json database > > > > > > 2. I started the drill console 'drill-embedded' > > > > > > 3. I then use the browser admin interfaced to enable MongoDB > > storage > > > > > engine > > > > > > 4. I followed the steps and was able to use zips collection > > > > > > 5. I was able to do count > > > > > > 6. I can't do queries for some reason: > > > > > > > > > > > > 0: jdbc:drill:zk=local> show Tables; > > > > > > +---------------+-------------+ > > > > > > | TABLE_SCHEMA | TABLE_NAME | > > > > > > +---------------+-------------+ > > > > > > | mongo.zips | zips | > > > > > > +---------------+-------------+ > > > > > > 1 row selected (0.131 seconds) > > > > > > 0: jdbc:drill:zk=local> alter system set > > > > > > `store.mongo.read_numbers_as_double` = true; > > > > > > +-------+----------------------------------------------+ > > > > > > | ok | summary | > > > > > > +-------+----------------------------------------------+ > > > > > > | true | store.mongo.read_numbers_as_double updated. | > > > > > > +-------+----------------------------------------------+ > > > > > > 1 row selected (0.088 seconds) > > > > > > 0: jdbc:drill:zk=local> select * from zips limit 10; > > > > > > Error: SYSTEM ERROR: IllegalArgumentException: Incoming > endpoints 1 > > > is > > > > > > greater than number of chunks 0 > > > > > > > > > > > > > > > > > > [Error Id: 4d8b145f-08bc-43a1-8156-b60b7c91fbc3 on > > > > cluster-server:31010] > > > > > > (state=,code=0) > > > > > > 0: jdbc:drill:zk=local> select count(*) from zips; > > > > > > +---------+ > > > > > > | EXPR$0 | > > > > > > +---------+ > > > > > > | 29353 | > > > > > > +---------+ > > > > > > 1 row selected (0.119 seconds) > > > > > > 0: jdbc:drill:zk=local> > > > > > > > > > > > > Please help. > > > > > > > > > > > > Thank you. > > > > > > Xiao > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > > > Kamesh. > > > > > > > > > > > > > > > > > > > > > -- > > > Kamesh. > > > > > > > > > -- > Kamesh. >
