[Bioperl-l] error running load_seqdatabase.pl

Hilmar Lapp hlapp at gmx.net
Fri Dec 30 19:33:19 EST 2005


I suspect it's either a (out-of-)memory error or a binary 
incompatibility.

You haven't yet run the bioperl-db test suite, have you? Please don't 
ask for more help before you haven't done that.

Also, you do remember my earlier answer to a similar report from you, 
right?
http://portal.open-bio.org/pipermail/bioperl-l/2005-December/020389.html

The two first things to do is to check whether you run out of shared 
memory and to run the tests and see whether the location where the 
process gets killed can be narrowed down. If it's not a bug (which it 
is unlikely to be - but the tests will provide more evidence) then the 
next step would be to recompile your own perl, DBI and DBD::Pg, and 
PostgreSQL, in that order.

	-hilmar

On Dec 30, 2005, at 4:13 PM, Angshu Kar wrote:

> oops! he is not present now!
> anyways, if you have some time could you please if there are any 
> reasons for the script to get killed with the .gbk file?
>
> Thanks,
> Angshu
>
> On 12/30/05, Hilmar Lapp <hlapp at gmx.net> wrote:
>> actually in a config file, and the default installation of Postgres
>> makes this config file unreadable for every one except the pg-admin
>> user chosen at installation time.
>>
>>         -hilmar
>>
>> On Dec 30, 2005, at 3:15 PM, Angshu Kar wrote:
>>
>> > I've no clue about my Pg setup! It has been done by the sysadmin! :(
>> > Could you please tell me the commands and I can run and let you know
>> > the reports?
>> >
>> > thanks,
>> > angshu
>> >
>> > On 12/30/05, Hilmar Lapp <hlapp at gmx.net> wrote:
>> >>
>> >> Also, you haven't reported your hw setup. In particular, how much
>> >> memory and swap do you have, is Pg running locally (your command 
>> line
>> >> suggests it does), and how much memory have you allocated to Pg?
>> >>
>> >> -hilmar
>> >>
>> >> On Dec 30, 2005, at 3:01 PM, Angshu Kar wrote:
>> >>
>> >> > Yes Hilmar its indeed a full sequence of chromosome. I got it 
>> from
>> >> > ftp://ftp.ncbi.nih.gov/genomes/Arabidopsis_thaliana/CHR_V/
>> >> > I tried with the .faa and .fna files and they ran successfully.
>> >> > Also I missed a line (may be relevant) in the debug o/p:
>> >> > no adaptor found for class Bio::Annotation::TypeManager
>> >> >
>> >> > Thanks,
>> >> > Angshu
>> >> >
>> >> >
>> >> > On 12/30/05, Hilmar Lapp < hlapp at gmx.net> wrote:
>> >> >> special about this one? The file name lets me guess the input 
>> is a
>> >> >> contig or even a full chromosome? Is the input on which the 
>> process
>> >> >> wasn't killed of the same nature? Are the sequences of 
>> comparable
>> >> or
>> >> >> vastly different lengths? Is there only a single contig in the
>> >> >> offending input? Can you reproduce the problem on an empty 
>> database
>> >> >> with input containing a single entry? If so, let me know where I
>> >> can
>> >> >> get that file from and I'll try to reproduce it locally.
>> >> >>
>> >> >> -hilmar
>> >> >>
>> >> >> On Dec 30, 2005, at 2:29 PM, Angshu Kar wrote:
>> >> >>
>> >> >> > Any clues why my load_seqdatabase.pl (./load_seqdatabase.pl
>> >> >> > --dbname=USBA --dbuser=postgres --format=genbank --driver=Pg
>> >> --debug
>> >> >> > NC_003076.gbk)
>> >> >> > is getting killed after the follwoing step?
>>  >> >> >
>> >> >> >Using Bio::DB::BioSQL::Pg::BiosequenceAdaptorDriver as driver 
>> peer
>> >> >> > for Bio::DB::BioSQL::BiosequenceAdaptor
>> >> >> > preparing SELECT statement: SELECT bioentry_id FROM 
>> biosequence
>> >> >> WHERE
>> >> >> > bioentry_id = ?
>> >> >> >
>> >> >> > Killed
>> >> >> >
>> >> >> > I'm using a linux box. This didn't happen for a different i/p
>> >> file!
>> >> >> >
>> >> >> > Thanks,
>> >> >> > Angshu
>> >> >> >
>> >> >> >
>> >> >> > On 12/29/05, Angshu Kar < angshu96 at gmail.com> wrote: I'm sorry
>> >> that
>> >> >> I
>> >> >> > missed it! Thanks for pointing it out.
>> >> >> >> Also thanks Hilmar for the update.
>> >> >> >>
>> >> >> >> Angshu
>> >> >> >>
>> >> >> >>
>> >>  >> >> On 12/29/05, Torsten Seemann <
>> >> >> torsten.seemann at infotech.monash.edu.au
>> >> >> >> > wrote:> I'm getting the following error while trying to 
>> run :
>> >> >> >>> > ./load_seqdatabase.pl -host localhost -dbname USBA -dbuser
>> >> >> >>> postgres -format
>> >> >> >>> > genbank NC_003076.gbk
>> >> >> >>> > MSG: failed to open connection: Can't connect to local 
>> MySQL
>> >> >> >>> server through
>> >> >> >>> > socket '/var/lib/mysql/mysql.sock' (2)
>> >> >> >>> > But I've a postgreSQL db and not a MySQL one...could 
>> anyone
>> >> >> please
>> >>  >> >>> guide me
>> >> >> >>> > troubleshoot this?
>> >> >> >>>
>> >> >> >>> Did you read the documentation for the load_seqdatabase.pl
>> >> script?
>> >> >> >>>
>> >> >> >>> It's in the script itself!
>> >> >> >>>
>> >> >> >>> =item --driver $driver
>> >> >> >>> the DBI driver name for the RDBMS e.g., mysql, Pg, or Oracle
>> >> >> >>> [mysql]
>> >> >> >>>
>> >> >> >>> The default DBI driver is [mysql].
>> >> >> >>>
>> >> >> >>> I think you want the option "-driver Pg".
>> >> >> >>>
>> >> >>>>> --
>> >> >> >>> Torsten Seemann
>> >> >> >>> Victorian Bioinformatics Consortium, Monash University,
>> >> Australia
>> >> >> >>> http://www.vicbioinformatics.com/
>> >> >> >>
>> >> >> >>
>> >> >> --
>> >> >> -------------------------------------------------------------
>> >> >> Hilmar Lappemail: lapp at gnf.org
>> >> >> GNF, San Diego, Ca. 92121phone: +1-858-812-1757
>> >> >> -------------------------------------------------------------
>> >> >>
>> >> >>
>> >> >
>> >> >
>> >> --
>> >> -------------------------------------------------------------
>> >> Hilmar Lappemail: lapp at gnf.org
>> >> GNF, San Diego, Ca. 92121phone: +1-858-812-1757
>> >> -------------------------------------------------------------
>> >>
>> >>
>> >>
>> --
>> -------------------------------------------------------------
>> Hilmar Lapp                            email: lapp at gnf.org
>> GNF, San Diego, Ca. 92121              phone: +1-858-812-1757
>> -------------------------------------------------------------
>>
>>
>>
-- 
-------------------------------------------------------------
Hilmar Lapp                            email: lapp at gnf.org
GNF, San Diego, Ca. 92121              phone: +1-858-812-1757
-------------------------------------------------------------





More information about the Bioperl-l mailing list