Our OOM was being caused by a damaged sequence data file. We had assumed
that the sequence files had checksums, which appears to be in correct.
The deserializer was reading a bad length out of the file and trying to
allocate 4gig of ram.
- question on Hadoop configuration for non cpu intensive... Jason Venner
- Re: question on Hadoop configuration for non cpu ... Ted Dunning
- Re: question on Hadoop configuration for non ... Jason Venner
- Re: question on Hadoop configuration for ... Ted Dunning
- RE: question on Hadoop configuration ... Devaraj Das
- Re: question on Hadoop configura... Jason Venner
- Re: question on Hadoop confi... Eric Baldeschwieler
- Re: question on Hadoop c... Jason Venner
- RE: question on Hadoop configura... Runping Qi
- Re: question on Hadoop configuration for non cpu ... Rui Shi