Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Plain View
Hadoop >> mail # user >> Map tasks processing some files multiple times


+
David Parks 2012-12-06, 06:15
+
Raj Vishwanathan 2012-12-06, 06:45
+
David Parks 2012-12-06, 07:36
+
Hemanth Yamijala 2012-12-06, 08:25
Copy link to this message
-
RE: Map tasks processing some files multiple times
I'm using multiple inputs because I actually have another type of input with
a different mapper, a single, unrelated file, that I omitted from this
discussion for simplicity. The basic formula is: read in a single database
of existing items, read in a bunch of catalogs of items, then merge and toss
like a salad (in a few map/reduce steps that follow).

 

 

From: Hemanth Yamijala [mailto:[EMAIL PROTECTED]]
Sent: Thursday, December 06, 2012 9:44 PM
To: [EMAIL PROTECTED]
Subject: Re: Map tasks processing some files multiple times

 

Glad it helps. Could you also explain the reason for using MultipleInputs ?

 

On Thu, Dec 6, 2012 at 2:59 PM, David Parks <[EMAIL PROTECTED]> wrote:

Figured it out, it is, as usual, with my code. I had wrapped TextInputFormat
to replace the LongWritable key with a key representing the file name. It
was a bit tricky to do because of changing the generics from <LongWritable,
Text> to <Text, Text> and I goofed up and mis-directed a call to
isSplittable, which was causing the issue.

 

It now works fine. Thanks very much for the response, it gave me pause to
think enough to work out what I had done.

 

Dave

 

 

From: Hemanth Yamijala [mailto:[EMAIL PROTECTED]]
Sent: Thursday, December 06, 2012 3:25 PM
To: [EMAIL PROTECTED]
Subject: Re: Map tasks processing some files multiple times

 

David,

 

You are using FileNameTextInputFormat. This is not in Hadoop source, as far
as I can see. Can you please confirm where this is being used from ? It
seems like the isSplittable method of this input format may need checking.

 

Another thing, given you are adding the same input format for all files, do
you need MultipleInputs ?

 

Thanks

Hemanth

 

On Thu, Dec 6, 2012 at 1:06 PM, David Parks <[EMAIL PROTECTED]> wrote:

I believe I just tracked down the problem, maybe you can help confirm if
you're familiar with this.

 

I see that FileInputFormat is specifying that gzip files (.gz extension)
from s3n filesystem are being reported as splittable, and I see that it's
creating multiple input splits for these files. I'm mapping the files
directly off S3:

 

       Path lsDir = new
Path("s3n://fruggmapreduce/input/catalogs/linkshare_catalogs/*~*");

       MultipleInputs.addInputPath(job, lsDir,
FileNameTextInputFormat.class, LinkShareCatalogImportMapper.class);

 

I see in the map phase, based on my counters, that it's actually processing
the entire file (I set up a counter per file input). So the 2 files which
were processed twice had 2 splits (I now see that in some debug logs I
created), and the 1 file that was processed 3 times had 3 splits (the rest
were smaller and were only assigned one split by default anyway).

 

Am I wrong in expecting all files on the s3n filesystem to come through as
not-splittable? This seems to be a bug in hadoop code if I'm right.

 

David

 

 

From: Raj Vishwanathan [mailto:[EMAIL PROTECTED]]
Sent: Thursday, December 06, 2012 1:45 PM
To: [EMAIL PROTECTED]
Subject: Re: Map tasks processing some files multiple times

 

Could it be due to spec-ex? Does it make a diffrerence in the end?

 

Raj

 
  _____  
From: David Parks <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Sent: Wednesday, December 5, 2012 10:15 PM
Subject: Map tasks processing some files multiple times

 

I've got a job that reads in 167 files from S3, but 2 of the files are being
mapped twice and 1 of the files is mapped 3 times.

 

This is the code I use to set up the mapper:

 

       Path lsDir = new
Path("s3n://fruggmapreduce/input/catalogs/linkshare_catalogs/*~*");

       for(FileStatus f : lsDir.getFileSystem(getConf()).globStatus(lsDir))
log.info("Identified linkshare catalog: " + f.getPath().toString());

       if( lsDir.getFileSystem(getConf()).globStatus(lsDir).length > 0 ){

              MultipleInputs.addInputPath(job, lsDir,
FileNameTextInputFormat.class, LinkShareCatalogImportMapper.class);

       }

 

I can see from the logs that it sees only 1 copy of each of these files, and
correctly identifies 167 files.

 

I also have the following confirmation that it found the 167 files
correctly:

 

2012-12-06 04:56:41,213 INFO
org.apache.hadoop.mapreduce.lib.input.FileInputFormat (main): Total input
paths to process : 167

 

When I look through the syslogs I can see that the file in question was
opened by two different map attempts:

 

./task-attempts/job_201212060351_0001/attempt_201212060351_0001_m_000005_0/s
yslog:2012-12-06 03:56:05,265 INFO
org.apache.hadoop.fs.s3native.NativeS3FileSystem (main): Opening
's3n://fruggmapreduce/input/catalogs/linkshare_catalogs/linkshare~CD%20Unive
rse~85.csv.gz' for reading

./task-attempts/job_201212060351_0001/attempt_201212060351_0001_m_000173_0/s
yslog:2012-12-06 03:53:18,765 INFO
org.apache.hadoop.fs.s3native.NativeS3FileSystem (main): Opening
's3n://fruggmapreduce/input/catalogs/linkshare_catalogs/linkshare~CD%20Unive
rse~85.csv.gz' for reading

 

This is only happening to these 3 files, all others seem to be fine. For the
life of me I can't see a reason why these files might be processed multiple
times.

 

Notably, map attempt 173 is more map attempts than should be possible. There
are 167 input files (from S3, gzipped), thus there should be 167 map
attempts. But I see a total of 176 map tasks.

 

Any thoughts/ideas/guesses?