What's the rest of the error message? Is fsck ok? Using the hadoop shell, can you read one of those files that's error'ing?
And more importantly, did you check the datanode log around the same timestamp? Thx, J-D On Mon, Mar 21, 2011 at 6:07 PM, Vivek Krishna <[email protected]> wrote: > I keep getting this error very often. > > java.io.IOException: java.io.IOException: Could not seek > StoreFileScanner[HFileScanner for reader reader=hdfs://e > > I have a 30 node cluster with several writers writing data. Once the write > is done and I run rowcounter job to count the records, I face the above > error. How do I fix this? > > Viv >
