Oh.  Yeah.  Makes sense.  We used to bundle the native libs but we
seem to have dropped them.  We should add them back?
St.Ack

On Tue, Jan 11, 2011 at 3:24 PM, Matt Corgan <[email protected]> wrote:
> Turns out this is what happens if you don't have the native libraries set up
> correctly.  The data still gets compressed using the pure java codec, but it
> doesn't cache the codec and gives you a warning each time it creates it for
> each block.
>
>
> On Mon, Jan 10, 2011 at 2:41 PM, Stack <[email protected]> wrote:
>
>> Thats a little silly.  That the message is INFO level is probably
>> small potatoes when doing a mapreduce job but in our case with lots of
>> file openings, it turns into a little log storm.
>>
>> I suppose you'll need to disable it.  Set log level to WARN on
>> org.apache.hadoop.io.compress?
>>
>> This might help you making the change:
>> http://wiki.apache.org/hadoop/Hbase/FAQ#A5
>>
>> St.Ack
>>
>> On Mon, Jan 10, 2011 at 9:46 AM, Matt Corgan <[email protected]> wrote:
>> > I'm trying to use GZIP compression but running into a logging problem.
>>  It
>> > appears that every time a block is compressed it logs the following:
>> >
>> > 2011-01-10 12:40:48,407 INFO org.apache.hadoop.io.compress.CodecPool: Got
>> > brand-new compressor
>> > 2011-01-10 12:40:48,414 INFO org.apache.hadoop.io.compress.CodecPool: Got
>> > brand-new compressor
>> > 2011-01-10 12:40:48,420 INFO org.apache.hadoop.io.compress.CodecPool: Got
>> > brand-new compressor
>> > 2011-01-10 12:40:48,426 INFO org.apache.hadoop.io.compress.CodecPool: Got
>> > brand-new compressor
>> > 2011-01-10 12:40:48,431 INFO org.apache.hadoop.io.compress.CodecPool: Got
>> > brand-new compressor
>> > 2011-01-10 12:40:48,447 INFO org.apache.hadoop.io.compress.CodecPool: Got
>> > brand-new compressor
>> > 2011-01-10 12:40:48,453 INFO org.apache.hadoop.io.compress.CodecPool: Got
>> > brand-new compressor
>> >
>> > Same for decompression.  It's logging that 150 times per second during a
>> > major compaction which pretty much renders the logs useless.  I assume
>> other
>> > people are not having this problem, so did we accidentally enable that
>> > logging somehow?
>> >
>> > Thanks,
>> > Matt
>> >
>>
>

Reply via email to