[Bioclusters] mpirun:

P Sravan Kumar. PAsst sravan at www.cdfd.org.in
Sun May 22 07:28:34 EDT 2005


Dear Sir, Thanks for your kind assitance. I have initiated the mpiblast
successfully yesterday. While running I have faced some error message.
It is as follows:
1. Process:
------------
NCBI: /apps/Software/NCBI-2.2.10-5.x86_64.rpm
mpiblast: /apps/Software/mpiblast-1.3.0-2LAM.x86_64.rpm
.ncbirc:
        [NCBI]
        Data=/apps/Databases/ncbi/data ( In this directory all the BLOSUM
and other ncbi data is there)
        [BLAST]
        BLASTDB=/apps/Databases/blastdb ( Its aimed to database created by
mpiformatdb at /apps/Databases/blastdb)
        BLASTMAT=/apps/Databases/ncbi/data ( It also to ncbidata).
        [mpiBLAST]
        Shared=/apps/Databases/blastdb ( full permission)
        Local=/tmp/mpiblast(full permission)
-----------------------------------------------------
in /home/sravan:
        mpiformatdb -N 250 -i /apps/Databases/silkbase/scaffolds(368MB) -p
F : has created a dataabase in /apps/Database/blastdb
        Then it has given: it has created a database with scaffolds(368MB)
- 250 fragments each with 2MB.

Script:
        /usr/local/lam/bin/mpirun -np 3 /usr/local/bin/mpiblast -p blastn
-i <input> -d <db> -o mpiblast.out

Error Message:
-------------
The sequence file has 1000 sequences of different sizes and as whole has
490kb. When I started the script , it ran properly upto 90 or 100 se
quences. Then also it continued. But during the process it has giving the
message serially:
[Blastall] ERROR: ncbiapi [000.000]  ObjMgrNextAvailEntityID failed with
idx 2048.

Kindly assist me in resolving this problem.

Thank you very much.
Regards.
Sravan.
=============================================
P.Sravana Kumar.
Laboratory of Molecular Genetics.
Center for DNA Fingerprinting and Diagnostics.
ECIL road, Nacharam.
Hyderabad 500 076.
Ph: 91-40-27151344-1102.
Ph: 91-40-27151344-1209.
e-mail: sravan at cdfd.org.in,sravan_mar at yahoo.com
==============================================



More information about the Bioclusters mailing list