Details
-
Sub-task
-
Status: Resolved
-
Major
-
Resolution: Done
-
None
-
None
-
None
-
None
Description
It looks like it might be a case of the starving-out-the-queue issue, but this time a valid query cannot run even in absence of errors.
It could also be something else (reducers being stuck?)
I ran on 10 nodes... with mode == map, query finishes.
With reducers in LLAP, it's stuck at this stage:
Map 1 .......... llap SUCCEEDED 31 31 0 0 0 0 Map 13 ......... llap SUCCEEDED 1 1 0 0 0 0 Map 14 ......... llap SUCCEEDED 1 1 0 0 0 0 Map 15 ......... llap SUCCEEDED 1 1 0 0 1 0 Map 5 .......... llap SUCCEEDED 31 31 0 0 0 00 Map 6 .......... llap SUCCEEDED 1 1 0 0 0 0 Map 7 .......... llap SUCCEEDED 1 1 0 0 0 0 Map 8 .......... llap SUCCEEDED 21 21 0 0 0 0 Map 9 llap RUNNING 91 0 91 0 0 00 Reducer 10 llap INITED 1 0 0 1 0 0 Reducer 11 llap INITED 1 0 0 1 0 0 Reducer 12 uber INITED 1 0 0 1 0 0 Reducer 3 llap RUNNING 13 0 13 0 0 0 Reducer 4 llap INITED 2 0 0 2 0 0