Hide Forgot
Description of problem: All map-reduce jobs are failing due to the new changed pathinfo format. regex in GlusterFSXattr.java needs to be updated to get new pathinfo format. Version-Release number of selected component (if applicable): master How reproducible: always Steps to Reproduce: 1. Run any map-reduce jobs with the master. Actual results: Fails with following exceptions. 12/02/24 07:00:14 INFO mapred.FileInputFormat: Total input paths to process : 1 java.lang.IllegalArgumentException: No enum const class org.apache.hadoop.fs.glusterfs.GlusterFSXattr$LAYOUT.DSRPPPP at java.lang.Enum.valueOf(Enum.java:214) at org.apache.hadoop.fs.glusterfs.GlusterFSXattr$LAYOUT.valueOf(GlusterFSXattr.java:35) at org.apache.hadoop.fs.glusterfs.GlusterFSXattr.getHints(GlusterFSXattr.java:246)
Sent the updated jar (in email). Please test it out and let me know. I'll send it out for review then.
With the new jar file I'm getting following error. 12/02/24 12:38:51 INFO mapred.FileInputFormat: Total input paths to process : 1 brick not of format hostname:path java.io.IOException: Error getting hostname from brick at org.apache.hadoop.fs.glusterfs.GlusterFSXattr.brick2host(GlusterFSXattr.java:49) at org.apache.hadoop.fs.glusterfs.GlusterFSXattr.getHints(GlusterFSXattr.java:346) at org.apache.hadoop.fs.glusterfs.GlusterFSXattr.getPathInfo(GlusterFSXattr.java:75)
CHANGE: http://review.gluster.com/2844 (hdfs/plugin: Fix regex for parsing pathinfo string) merged in master by Vijay Bellur (vijay)