Hi Guys!

I ingest the data about 20gb(4 years of sales data) everything was working fine.

I used this command: bin/post-index-task --file /home/umer/sale-csv.json --url http://localhost:8090

Data has been indexed successfully

showing on screen

Task finished with status: SUCCESS

Completed indexing data for sale_sm_theta. Now loading indexed data onto the cluster…

sale_sm_theta is 0.0% finished loading…

sale_sm_theta is 0.0% finished loading…

sale_sm_theta is 0.0% finished loading…

sale_sm_theta is 0.0% finished loading…

sale_sm_theta is 0.0% finished loading…

sale_sm_theta is 0.0% finished loading…

sale_sm_theta is 0.0% finished loading…

sale_sm_theta is 0.0% finished loading…

.

.

.

.

.

.

.

.

.

.

.

.

Traceback (most recent call last):

File “/home/umer/apache-druid-0.14.2-incubating/bin/post-index-task-main”, line 174, in

main()

File “/home/umer/apache-druid-0.14.2-incubating/bin/post-index-task-main”, line 171, in main

await_load_completion(args, datasource, load_timeout_at)

File “/home/umer/apache-druid-0.14.2-incubating/bin/post-index-task-main”, line 132, in await_load_completion

raise Exception("{0} was not loaded in time!".format(datasource))

**Exception: sale_sm_theta was not loaded in time! **

my question is why it takes more time than index?

And how can I load my index data. without resubmitting the task again?

Hi Umar,

I did test "post-index-task" for Wikipedia test data and it worked fine.

I suggest checking the middle-manager logs to make sure that the task is successful. Then you can investigate historical logs for any errors in loading the segments (org.apache.druid.server.coordination.BatchDataSegmentAnnouncer).

Thanks,

Hemanth