confirmation required regarding the BACKUP-NAMENODE-BEHAVIOR when it is performing checkPoint

[sandeep <sa...@huawei.com>]

Hi,

 

Please help me in confirming the behavior of BNN when it is performing
checkPoint.

 

Scenario:

=======

1)Start Namenode.and BackupNameNode

2)Start writing files to HDFS continuously ....Say after 5 min
BackupNameNode is invoking doCheckpoint . then following exception is thrown
at BNN side

 

Exception:

========

  error:java.io.IOException:java.lang.IllegalStateException:!!! WARNING !!!
File system changes are not persistent .No journal streams

 at
org.apachel.hadoop.hdfs.server.namenode.FSEditLog.logEdit(FSEditLog.java:157
0)

 

Please find my Analysis for the above exception:

===================================

-->While checkpointing BNN will call startCheckpoint  on namenode and it
will get NamenodeCommand

  then saveCheckpoint is called at BNN as part of this it will clear the
editStreams  in order to rename current dir to latestcheckpoint.tmp ,some
operations are performed on BNN  like   renaming current dir to
latestcheckpoint.tmp ...etc ..then it will populate the editStreams

 

At this time if any dfsclient is trying to write data to hdfs the above
exception is thrown .

 

 

PLEASE LET ME KNOW ABOUT THIS BEHAVIOUR ,AND HOW THE ABOVE EXCEPTION CAN BE
HANDLED

 

Thanks&Regards

sandeep

 

 

 

****************************************************************************
***********
This e-mail and attachments contain confidential information from HUAWEI,
which is intended only for the person or entity whose address is listed
above. Any use of the information contained herein in any way (including,
but not limited to, total or partial disclosure, reproduction, or
dissemination) by persons other than the intended recipient's) is
prohibited. If you receive this e-mail in error, please notify the sender by
phone or email immediately and delete it!