Looks like this is your error...
Caused by: org.apache.hadoop.mapreduce.lib.input.InvalidInputException: Input
path does not exist:
hdfs://tiberiumtan-nn1.tan.ygrid.yahoo.com:8020/user/gdfm/implicit/implicit/flow-all-giraph
at
org.apache.hadoop.mapreduce.lib.input.FileInputFormat.listStatus(FileInputFormat.java:235)
at
org.apache.hadoop.mapreduce.lib.input.FileInputFormat.getSplits(FileInputFormat.java:252)
at
org.apache.giraph.lib.TextVertexInputFormat.getSplits(TextVertexInputFormat.java:120)
at
org.apache.giraph.graph.BspServiceMaster.generateInputSplits(BspServiceMaster.java:229)
Avery
On 5/24/12 10:36 AM, Gianmarco De Francisci Morales wrote:
Hi,
I have a problem in launching a large Giraph job.
The same code works for a small number of workers, but when I try to
get more I get the stack trace attached.
I assume it must be some kind of inter machine communication problem.
Is it a know issue by any chance?
Thanks,
--
Gianmarco
