Dennis Kubes wrote:
I believe the percentage complete is set in hadoop, in the TaskInProgress.recomputeProgressMethod() and then lines 570-595 in JobInProgress.updateTaskStatus.
Correct - this actually comes down the way from reading the current FileSplit, i.e. the part of the input fetchlist. When this reading process is completed, the percentage is set to 100%, even though a lot of URLs could be still queued and waiting to be fetched.
-- Best regards, Andrzej Bialecki <>< ___. ___ ___ ___ _ _ __________________________________ [__ || __|__/|__||\/| Information Retrieval, Semantic Web ___|||__|| \| || | Embedded Unix, System Integration http://www.sigram.com Contact: info at sigram dot com
