problems under same hosts and different ip addresses

hi, hadoopers. Recently, we've moved our clusters to another idc center. We keep the same host-names, but, they have now different ip addresses. Without any configuration changes, we got the following error after starting cluster. 13.110.239.218 <-- old ip 13.271.6.54 <-- new ip 2013-02-20 1

Re: is there a standard way to set a custom FileOutputCommitter?

> http://hadoop.apache.org/common/docs/r0.20.0/api/org/apache/hadoop/mapred/JobConf.html#setOutputCommitter(java.lang.Class) > > On Sun, May 8, 2011 at 9:40 AM, JunYoung Kim wrote: >> Hi, >> >> I want to set my custom (overried) FileOutputCommitter because it prov

is there a standard way to set a custom FileOutputCommitter?

Hi, I want to set my custom (overried) FileOutputCommitter because it provides the method named 'rename'. This method has a role to move an output dir in temp output to a real output dir. but, it checks output file is existed or not before moving it. this logic is not proper in my case. I w

does counters go the performance down seriously?

hi, this linke is about hadoop usage for the good practices. http://developer.yahoo.com/blogs/hadoop/posts/2010/08/apache_hadoop_best_practices_a/ by Arun C Murthy if I want to use about 50,000 counters for a job, does it cause serious performance down?

is there a way to write rows sequentially against 60 reduce tasks?

hi, I run almost 60 ruduce tasks for a single job. if the outputs of a job are from part00 to part 59. is there a way to write rows sequentially by sorted keys? curretly my outputs are like this. part00) 1 10 12 14 part 01) 2 4 6 11 13 part 02) 3 5 7 8 9 but, my aim is to get the following

how to write outputs sequentially?

hi, I run almost 60 ruduce tasks for a single job. if the outputs of a job are from part00 to part 59. is there way to write rows sequentially by sorted keys? curretly my outputs are like this. part00) 1 10 12 14 part 01) 2 4 6 11 13 part 02) 3 5 7 8 9 but, my aim is to get the following r

Re: what's the differences between file.blocksize and dfs.blocksize in a job.xml?

hi, harsh. is there a way to put my file on a hdfs with another block size? usually, I did to copy my files to a hdfs. $> hadoop fs -copyFromLocal localFile hdfsFile do I need to put some another field to re-create in my command? thank 2011. 3. 13., 오후 5:42, Harsh J 작성: > Hello, > > On Wed

Re: How to count rows of output files ?

actually, a structure of output directories are quite complexed. A directory has 1, 2, 3 as output files B directory has 1, 2, 3, 4 as output files C directory has 1, 2, 3, 5 as output files structure of directories, simply 2011 |- A |- 1 | |- 2 | |- 3

Re: is there more smarter way to execute a hadoop cluster?

i, Harsh. >> >> I've already tried to do use tag to set it unmodifiable. >> but, my result is not different. >> >> *core-site.xml:* >> >> >> fs.default.name >> hdfs://localhost >> true >> >> >> >> other