I am trying to use snapshot+WALPlayer for HBase DR for our cluster in AWS. I am trying to do below to verify it, seems the new data is not being played into the new table. Anything wrong with my steps? 1. Populate TestTable using PeformanceEvaluation Tool 2. count the rows being written, 63277 row(s) 3. take a snapshot, then clone a table TestTable-cloned based on this snapshot. Count the row # and verified that is has same number of rows as TestTable 4. write more data to TestTable using PerformanceEvaluation 5. count the row of TestTable, which has more rows. 6. call WALPlayer: su tychang hbase org.apache.hadoop.hbase.mapreduce.WALPlayer /hbase/.logs TestTable-cloned 7. count the row of TestTable-cloned. Find the row count is not changed, still 63277 row(s) :(
I suspect that the WAL is not rolled yet, so WALPlayer cannot replay those data. So I populated more data to make sure the WAL log rolled.(I can see the count of hlogs increased by 10) But still, after running WALPlayer, the rowCount is not changed. Any idea? Can WALPlayer work with Snapshot? Thanks Tian-Ying On Fri, Feb 7, 2014 at 10:31 AM, Tianying Chang <[email protected]> wrote: > Hi, Lars > > Sure. I will come back and update the thread. > > Thanks > Tian-Ying > > > On Fri, Feb 7, 2014 at 10:17 AM, lars hofhansl <[email protected]> wrote: > >> Let me know how this works for you. >> I wrote that tool a while ago, but I ended up never actually using myself. >> >> -- Lars >> >> >> >> ________________________________ >> From: Tianying Chang <[email protected]> >> To: [email protected] >> Sent: Thursday, February 6, 2014 10:06 PM >> Subject: Re: WALPlayer? >> >> >> Never mind. Should use hbase command. :) >> >> Thanks >> Tian-Ying >> >> >> >> On Thu, Feb 6, 2014 at 9:53 PM, Tianying Chang <[email protected]> wrote: >> >> > Hi, folks >> > >> > I want to try the WALPlayer. But it complains not found. Am I running it >> > the wrong way? >> > >> > Thanks >> > Tian-Ying >> > >> > hadoop jar /tmp/hbase-0.94.17-SNAPSHOT.jar WALPlayer >> > >> > Unknown program 'WALPlayer' chosen. >> > >> > Valid program names are: >> > >> > CellCounter: Count cells in HBase table >> > >> > completebulkload: Complete a bulk data load. >> > >> > copytable: Export a table from local cluster to peer cluster >> > >> > export: Write table data to HDFS. >> > >> > import: Import data written by Export. >> > >> > importtsv: Import data in TSV format. >> > >> > rowcounter: Count rows in HBase table >> > >> > verifyrep: Compare the data from tables in two different clusters. >> > WARNING: It doesn't work for incrementColumnValues'd cells since the >> > timestamp is changed after being appended to the log. >> > >> > >
