hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sunita Arvind <>
Subject Help with hive debugger
Date Wed, 19 Jun 2013 17:32:40 GMT
Some of you might have seen the original issue that I am trying to resolve.
(select xxx from table_name ; shows null ) on the user mailing list.

I am trying to attach btrace to debug this inorder to see the value of the
token retrieved by the parser.

I need help regarding:
1. starting hive in debug mode:
hive> [sunita@node0hive --debug, recursive=y, port=7000,mainSuspend=y,
childSuspend=y =9009"
ERROR: Cannot load this JVM TI agent twice, check your java command line
for duplicate jdwp options.
Error occurred during initialization of VM
agent library failed to init: jdwp

I tried multiple port numbers but see the same error

2. I can connect btrace to hive using the pid of hive process.
btrace <pid> <btrace-script>
   But how do I get btrace talk to the actual map reduce job which will act
on the data. AFAIK, hive will spawn another JVM which may or may not be on
the same machine.

3. I know of btrace as a dynamic debugging tool. Which is quite good (I
haven't used it much though) for java applications. For this scenario, is
there a better tool?

Appreciate your help.


  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message