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

Switch to Threaded View
Accumulo >> mail # user >> Reduce task failing on job with error java.lang.IllegalStateException: Keys appended out-of-order


Copy link to this message
-
Re: Reduce task failing on job with error java.lang.IllegalStateException: Keys appended out-of-order
The point of bulk-ingest is that you can perform this work "out of band"
from Accumulo. You can perform the work "somewhere else" and just tell
Accumulo to bring files online. The only potential work Accumulo has to
do at that point is maintain the internal tree of files (merging and
splitting as the table is configured). Given that we have this massively
popular tool for performing distributed sorting (cough MapReduce cough),
I don't agree with your assertion.

If you don't want to be burdened with sorting output during the ingest
task, use live ingest (BatchWriters). For reasonable data flows, live
ingest tends to be faster; however, bulk ingest provides the ability to
scale to much larger flows of data while not tanking Accumulo.

On 12/6/12 9:15 AM, Chris Burrell wrote:
> Is this a limitation of the bulk ingest approach? Does the MapReduce
> job need to give the data to the AccumuloOutputFileFormat in
> a lexicographically-sorted manner? If so, is this not a rather big
> limitation of this approach, as you need to ensure your data comes in
> from your various data sources in a form such that the accumulo keys
> are then sorted.
>
> This seems to suggest that although the bulk ingest would be very
> quick, you would lose most of the time trying to sort and adapt the
> source files themselves in the MR job.
>
> Chris
>
>
>
> On 6 December 2012 14:08, William Slacum
> <[EMAIL PROTECTED]
> <mailto:[EMAIL PROTECTED]>> wrote:
>
>     Excuse me, 'col3' sorts lexicographically *after* 'col16'.
>
>
>     On Thu, Dec 6, 2012 at 9:07 AM, William Slacum
>     <[EMAIL PROTECTED]
>     <mailto:[EMAIL PROTECTED]>> wrote:
>
>         'col3' sorts lexicographically before 'col16'. you'll either
>         need to encode your numerics or zero pad them.
>
>
>         On Thu, Dec 6, 2012 at 9:03 AM, Andrew Catterall
>         <[EMAIL PROTECTED]
>         <mailto:[EMAIL PROTECTED]>> wrote:
>
>             Hi,
>
>
>             I am trying to run a bulk ingest to import data into
>             Accumulo but it is failing at the reduce task with the
>             below error:
>
>             java.lang.IllegalStateException: Keys appended
>             out-of-order.  New key
>             client@20121206123059@0014efca-d8e8-492e-83cb-e5b6b7c49f7a
>             foo:col3 [myVis] 9223372036854775807 false, previous key
>             client@20121206123059@0014efca-d8e8-492e-83cb-e5b6b7c49f7a
>             foo:col16 [myVis] 9223372036854775807 false
>
>             at
>             org.apache.accumulo.core.file.rfile.RFile$Writer.append(RFile.java:378)
>
>             Could this be caused by the order at which the writes are
>             being done?
>
>
>             *-- Background*
>
>             *
>             *
>
>             The input file is a tab separated file.  A sample row
>             would look like:
>
>             Data1 Data2    Data3    Data4    Data5 �             DataN
>
>             The map parses the data, for each row, into a Map<String,
>             String>.  This will contain the following:
>
>             Col1 Data1
>
>             Col2 Data2
>
>             Col3 Data3
>
>             �
>
>             ColN DataN
>
>
>             An outputKey is then generated for this row in the format
>             *client@timeStamp@randomUUID*
>
>             Then for each entry in Map<String, String> a
>             outputValue is generated in the format *ColN|DataN*
>
>             The outputKey and outputValue are written to Context.
>
>             This completes successfully, however, the reduce task fails.
>
>
>             My ReduceClass is as follows:
>
>             *public**static**class* ReduceClass
>             *extends* Reducer<Text,Text,Key,Value> {
>
>             *public**void* reduce(Text key, Iterable<Text> keyValues,
>             Context output) *throws* IOException, InterruptedException {
>
>             // for each value belonging to the key