Segmentation fault

173 views
Skip to first unread message

Yun-juan Chang

unread,
Mar 11, 2013, 3:11:54 PM3/11/13
to bgi-...@googlegroups.com
Hi,

When I use SOAPec, run KmerFreq_AR with space kmer, I keep getting the error message: Segmentation fault, but the first part of KmerFreq_AR without space kmer went through fine. Further, both KmerFreq_AR run  with/without space kmer completed successfully with my other set of data.

Therefore, I suspect this is data related, in this case, also space kmer related.  My reads data file is about 54 Gb, could this exceed the limit of SOAPec ? Or there maybe other explanation ?

Any help will be appreciated.  Thanks!

Janet
 

Justin Petyon

unread,
Oct 15, 2013, 1:06:58 PM10/15/13
to bgi-...@googlegroups.com
Hello,

Did anyone ever figure out what was causing the error? I have run into a similar problem. I am working with about 104G of data which is about 25X for the genome I am working with. I am working on a Ubuntu 12.04 64 system with 32G of ram. When I run this data through with out the -s option, everything seems to work. I get the following output.

Program start..
Program: KmerFreq_AR
Version: v2.0
Author: BGI-ShenZhen
CompileDate: Dec 17 2012 time: 10:56:25
Current time: Tue Oct 15 12:16:10 2013
Command line: KmerFreq_AR -t 8 -k 15 -p bullata FileList.txt

Begin to construct Kmer frequency table...
Kmer frequency table initialization completed
Run time: 0s.
Start to parse read file: W.1B.fq ...
8 threads were created!
Finished parse reads file: W.1B.fq
Run time: 193s.
Start to parse read file: W.2B.fq ...
8 threads were created!
Finished parse reads file: W.2B.fq
Run time: 387s.
Start to parse read file: W.1S.fq ...
8 threads were created!
Finished parse reads file: W.1S.fq
Run time: 389s.
Start to parse read file: W.2S.fq ...
8 threads were created!
Finished parse reads file: W.2S.fq
Run time: 389s.
Parsed all the reads files completed
Run time: 389s.
Start to combine forward and reverse strands of Kmers...
8 threads were created!
Complete to combine forward and reverse strands of Kmers!
Construction of Kmer frequency table completed!
Run time: 396s.
Start to output compressed kmer frequency file...
8 threads were created!
Complete to generate the kmer frequency compressed file!
Run time: 416s.
Start to generate kmer frequency statistics file...
Complete to generate the kmer frequency statistics file!
Run time: 416s.
Please check the peek position carefully.
Start to generate genome characters estimate-file...
Complete to generate the genome characters estimate-file!

All done!
Run time: 416s.

When I run the exact same setup with -s 2 option I get. 

Program start..
Program: KmerFreq_AR
Version: v2.0
Author: BGI-ShenZhen
CompileDate: Dec 17 2012 time: 10:56:25
Current time: Tue Oct 15 12:14:41 2013
Command line: KmerFreq_AR -t 8 -s 2 -k 15 -p bullata FileList.txt

Begin to construct Kmer frequency table...
Kmer frequency table initialization completed
Run time: 1s.
Start to parse read file: W.1B.fq ...
8 threads were created!

And I get: Segmentation fault (core dumped) from the standard error.

Is it possible that I am running into a memory problem? I have 32G memory and without the -s flag, SOAPec uses around 1G of memory. I do not know if it is a related problem, but I tried to track down the possibility that the problem was a result of too little memory and ran into another issue. Because the example data provided with SOAPdenovo was so small, I decided to try a few things out with it. When running multitreaded with -s option the program started thrashing at the "output compressed kmer frequency file" stage:

peyton@Cranium:~/Downloads/test$ KmerFreq_AR -t 8 -k 11 -s 4 -p Test TestList.txt
Program start..
Program: KmerFreq_AR
Version: v2.0
Author: BGI-ShenZhen
CompileDate: Dec 17 2012 time: 10:56:25
Current time: Mon Oct 14 15:35:21 2013
Command line: KmerFreq_AR -t 8 -k 11 -s 4 -p Test TestList.txt

Begin to construct Kmer frequency table...
Kmer frequency table initialization completed
Run time: 0s.
Start to parse read file: test_PE1.fa ...
8 threads were created!
Finished parse reads file: test_PE1.fa
Run time: 0s.
Start to parse read file: test_PE2.fa ...
8 threads were created!
Finished parse reads file: test_PE2.fa
Run time: 0s.
Parsed all the reads files completed
Run time: 0s.
Start to combine forward and reverse strands of Kmers...
8 threads were created!
Complete to combine forward and reverse strands of Kmers!
Construction of Kmer frequency table completed!
Run time: 0s.
Start to output compressed kmer frequency file...
8 threads were created!
^C

If, however, I run essentially the same setup single threaded it runs like a champ.

Program start..
Program: KmerFreq_AR
Version: v2.0
Author: BGI-ShenZhen
CompileDate: Dec 17 2012 time: 10:56:25
Current time: Mon Oct 14 17:07:39 2013
Command line: KmerFreq_AR -k 11 -s 4 -p Test TestList.txt

Begin to construct Kmer frequency table...
Kmer frequency table initialization completed
Run time: 0s.
Start to parse read file: test_PE1.fa ...
1 threads were created!
Finished parse reads file: test_PE1.fa
Run time: 1s.
Start to parse read file: test_PE2.fa ...
1 threads were created!
Finished parse reads file: test_PE2.fa
Run time: 1s.
Parsed all the reads files completed
Run time: 1s.
Start to combine forward and reverse strands of Kmers...
1 threads were created!
Complete to combine forward and reverse strands of Kmers!
Construction of Kmer frequency table completed!
Run time: 1s.
Start to output compressed kmer frequency file...
1 threads were created!
Complete to generate the kmer frequency compressed file!
Run time: 2s.
Start to generate kmer frequency statistics file...
Complete to generate the kmer frequency statistics file!
Run time: 2s.
Please check the peek position carefully.
Start to generate genome characters estimate-file...
Complete to generate the genome characters estimate-file!

All done!
Run time: 2s.

I tried running the large data set single threaded, but it still failed in the same way. Thank you in advance for the help.

Justin

Steven VB

unread,
May 20, 2014, 7:14:18 AM5/20/14
to bgi-...@googlegroups.com
Hello Justin,

I have the exact same problem.. did you solve it by compiling the new version?

Thanks,
Steven

Steven VB

unread,
May 20, 2014, 12:49:20 PM5/20/14
to bgi-...@googlegroups.com
Hi,

could there be something wrong with the format of my fastq files?

@FCC4BKEACXX:7:1101:1201:2157#GAATCGAT/1
ATTTTTCTTGAAAAAAGTATGAAATTTTACTAAAATTCGATAAAAAAATGTAAAAAATTTCGTAAAAATTCAAATTTTTAACAATATTTTATTACCAGAT
+
bbbeeeeegggggiiiiiiiiiiiiiiiiiiiiiiiihiihiiiiiiiiihiiiiiigggggeeeedddddddccccccccccccccddddccdccccbc
@FCC4BKEACXX:7:1101:1149:2159#GAATCGAN/1
ACGNCAATGGTTTCCCCATAAAATTATCTATAAAAATACATTCAGTCCAATCACCCTTTAAACCAGTCCAATTTAAAAAAAAGTAAGACTATATTTCTAN
+
bbbBS`ccggeggiiiiiiiiiiiiiiiifhiiihhihihiiiiifhiihiiiiiiiiiihhiihihiiiihiihhgggeccc_bcccccccbcedddcB
@FCC4BKEACXX:7:1101:1348:2070#GAATCGAT/1
TCTNGAAAACTAAGACATGTTTTTGAAAACGGTCTACGGCATTAGATTCCCCATAAAATTTTCTATTAAAATACCTAAAGTCTAGTCACCCTTAAAACCA


Thanks,
Steven
Reply all
Reply to author
Forward
0 new messages