Configuring Hadoop for Failover

Introduction As of 0.20, Hadoop does not support automatic recovery in the case of aĀ NameNodeĀ failure. This is a well known and recognized single point of failure in Hadoop. Experience at Yahoo! shows thatĀ NameNodesĀ are more likely to fail due to misconfiguration, network issues, and bad behavior amongst clients than actual hardware problems. Out of fifteen grids…

Measuring Apache bandwidth from a bash script

So reason : double checking Sawmill and AWstats. (sawmill won..) echo “$(awk ‘{print $10}’ prd.logs | grep -v “-” | paste -sd + – | bc )/1024/1024/1024″ | bc result is in GB. nJoy;