problem with picard fixmateinfo

[Kotoro]

command used in a function defined in a BASH script:

java -Xmx$JAVA_RAM -Xms$JAVA_RAM -jar /UCHC/HPC/Everson_HPC/picard/bin/FixMateInformation.jar \
MAX_RECORDS_IN_RAM=$MRECORDS \
I=./GATK_prep/$1.addReadGroups \
O=./GATK_prep/$1.fixMateInfo

exception that resulted:

grep Exception qsub_logs/SKA091711TCBL5B.step2-6.*
qsub_logs/SKA091711TCBL5B.step2-6.err:Exception in thread "main" net.sf.samtools.util.RuntimeIOException: java.io.IOException: No space left on device
qsub_logs/SKA091711TCBL5B.step2-6.err:Caused by: java.io.IOException: No space left on device
qsub_logs/SKA091711TCBL5B.step2-6.err:##### ERROR MESSAGE: Couldn't read file /UCHC/Everson/Projects/Bladder/Pt5/B/SKA091711TCBL5B.4GATK.recal.bam because java.io.FileNotFoundException: SKA091711TCBL5B.4GATK.recal.bam (No such file or directory)
qsub_logs/SKA091711TCBL5B.step2-6.err:##### ERROR MESSAGE: Couldn't read file /UCHC/Everson/Projects/Bladder/Pt5/B/SKA091711TCBL5B.4GATK.recal.bam because java.io.FileNotFoundException: SKA091711TCBL5B.4GATK.recal.bam (No such file or directory)

I have recently freed almost 7 TB of data that we had been storing on this cluster's user share (which is approximately 84T in size and only 75% used). The run on this file is not producing enough data to warrant hitting any user-based quotas that I would not have already encountered previously if they had been in place. In fact I can create many duplicate copies of large files manually with no disk usage warnings.

And yet picard's tool claims it is running out of disk space. Whats wrong? How do I get past this error?

[FractalExpression]

I would explicitly declare a scratch directory, as I've had this issue in the past with FixMateInfo:

-Djava.io.tmpdir= /your_scratch_directory

Also, somewhat redundant, but just to be safe add:

TMP_DIR=/your_scratch_directory

Good luck.

[iyer0000]

even i have problem with this posts,please move to stack underflow