THIS IS A TEST INSTANCE. ALL YOUR CHANGES WILL BE LOST!!!!
...
| Code Block | ||||
|---|---|---|---|---|
| ||||
cd $MY$TRAF_SQROOTHOME/sql/scripts sqstart sqcheck |
...
If problems persist please review logs:
Start with the log file with the recent modification time and traverse from any event backwards to the root cause of the problem$TRAF_HOME$MY_SQROOT/sql/local_hadoop/*/log: Hadoop, HBase, and Hive logs.$MY$TRAF_SQROOTHOME/logs: Trafodion logs.- In Trafodion 2.1 TM has been improved to log all critical events into tm_<nid>.log
- Look for TM core files in directory /proc/sys/kernel/core_pattern or in $TRAF_HOME/sql/scripts
- Look at the stdout_<traf_component> files in $TRAF_HOME/sql/scripts
Manage Hadoop Environment
...
If you are doing Trafodion development on a pre-existing Hadoop distribution, then do the following:
- Distribution: Use the distribution management too. For example: Apache Ambari or Cloudera Manager.
- Regular Hadoop: Use the start/stop script for each Hadoop environment.
Local Hadoop
Use the following commands to manage the Hadoop environment.
| Command | Usage |
|---|---|
swstartall | Start the complete Hadoop environment. |
swstopall | Stops the complete Hadoop environment. |
swstatus | Checks the status of the Hadoop environment. |
swuninstall_local_hadoop | Removes the Hadoop installation. |
Downloading New Source Code
You need to do the following each time you download new source code.
| Code Block | ||
|---|---|---|
| ||
cd <Trafodion source directory> source ./env.sh cd $MY_SQROOT/etc # delete ms.env, if it exists rm ms.env cd $MY_SQROOT/sql/scripts sqgen |
| Code Block | ||||
|---|---|---|---|---|
| ||||
cd mysource/incubator-trafidion
source ./env.sh
cd $MY_SQROOT/etc
# delete ms.env, if it exists
rm ms.env
cd $MY_SQROOT/sql/scripts
sqgen |
...