I tried to access the same file from another machine using hadoop cli and am able to view the file contents. But still I did, try with another file and issue exists.
Could there be any problem due to impersonation? As I am running drill and zookeeper in one instance and the hadoop services in another machine? Will there be any problem with ports?(Though there is no firewall running in both the machines). Thanks, Malathi On Mon, 17 Aug 2015 at 21:35 Abhishek Girish <[email protected]> wrote: > Hey, > > I haven't seen this before - but a quick google search makes me think it's > an issue on Hadoop side (concurrency issues, permissions on file, stale > replicas, ...). Could you possibly create a copy of this file and try > again? > > Others can comment if they know better. > > -Abhishek > > On Mon, Aug 17, 2015 at 2:53 AM, Malathi <[email protected]> wrote: > > > Hi, > > > > I am trying to access HDFS data sources running in a remote instance from > > apache drill running in my laptop. But I get the following error : > > > > 2015-08-13 06:06:48,287 [Client-1] INFO > > o.a.d.j.i.DrillResultSetImpl$ResultsListener - [#2] Query failed: > > org.apache.drill.common.exceptions.UserRemoteException: SYSTEM ERROR: > > BlockMissingException: Could not obtain block: > > BP-67039680-172.17.0.1-1434643788667:blk_1073820202_79472 > > file=/tmp/test.json > > > > I am able to view the files via the datasource, but > > cannot issue a select * query. Can you please help? PFA the data source > > configuration. > > > > Thanks, > > Malathi > > >
