[Bioperl-l] Error loading sequence with load_seqdatabase.pl

Hilmar Lapp hlapp at gmx.net
Wed Jun 8 23:46:10 EDT 2005


I've never heard about Rocks Cluster as an OS ... 2GB seem plenty; even 
if you db is on the same machine it shouldn't be a problem (unless 
you've most of the memory to Pg's shared pool). You need to use a 
monitoring tool like 'top' to see where your resources go and whether 
something exhausts the memory.

If you write a simple script that creates a simple dummy table and 
inserts 1 million random rows, does that raise any problem? (Like, 
accumulates memory or even crashes too?)

	-hilmar

On Jun 9, 2005, at 11:39 AM, Duangdaow Kanhasiri wrote:

> The
>
> OS: Rocks Cluster v 3.3
> Total Memory: 2 GB
> DBD::Pg version: 1.42
> DBI version: 1.48
>
>
> --- Hilmar Lapp <hlapp at gmx.net> wrote:
>
>> What OS are you running this on? How much memory
>> have you got on the
>> machine on which you run the script, and on the
>> machine on which you
>> run the database? Are these the same or not? Which
>> version of DBI and
>> DBD::Pg?
>>
>> This hasn't been reported by anyone else really so I
>> suspect it's
>> either due to too limited memory, or a problem in
>> the DBD driver or in
>> the DBI compiled code. Can you watch the process
>> (using, e.g., top) and
>> see how fast it increases in memory consumption?
>> Since you can continue
>> when you restart it's not something specific to one
>> sequence that would
>> trigger the problem; rather it appears whenever you
>> have run through a
>> certain number of entries the process dies.
>>
>> 	-hilmar
>>
>> On Jun 8, 2005, at 7:43 PM, Duangdaow Kanhasiri
>> wrote:
>>
>>> Hi,
>>>
>>> I've used the bioperl script load_seqdatabase.pl
>> (came
>>> with the biosql' scripts) to load the bacterial
>>> sequence in genbank format(*.gbk) into PostgreSQL
>> 8.0
>>> database on Linux machine as:
>>>
>>> $perl load_seqdatabase.pl /export/Bacteria/*/*.gbk
>> &
>>>
>>> Where  under the /export/Bacteria/ path are the
>>> Bacteria's name path e.g. Acinetobacter_sp_ADP1
>> and
>>> the file name are like NC_006824.gbk.
>>>
>>> Previously it used to load some sequences in to
>> some
>>> tables in biosql database (count from table
>> bioentry)
>>>
>>> bioseq=# select count(*) from bioentry;
>>>  count
>>> -------
>>>    33
>>> (1 row)
>>>
>>>
>>> However, after a while it then stopped with the
>> the
>>> error:
>>>
>>> [1]+  Segmentation fault      perl
>> load_seqdatabase.pl
>>> /export/Bacteria/*/*.gbk &
>>>
>>> I then checked and removed the *.gbk file that
>> have
>>> already been loaded in to the table, leaving only
>> the
>>> unloaded ones and ran the scripted again.  It
>>> continued to work for some times and stopped
>> again.  I
>>> repeated the process several times until 173
>> sequences
>>> were loaded into the table:
>>>
>>> bioseq=# select count(*) from bioentry;
>>>  count
>>> -------
>>>    173
>>> (1 row)
>>>
>>> The program then stopped again and this time it
>>> wouldn't run anymore even I tried with only on
>> file.
>>> The error is still the same like:
>>>
>>> $ perl load_seqdatabase.pl
>>>
>>
> /export/Bacteria/Lactobacillus_johnsonii_NCC_533/NC_005362.gbk
>>> Segmentation fault
>>> $
>>>
>>> Now I couldn't load the rest of my sequences into
>> the
>>> database anymore.  I would be very apprecialed if
>> any
>>> one knows how to solve the "Segmentation fault"
>>> problem?
>>>
>>> Regards,
>>>
>>> Davina
>>>
>>>
>>> 		
>>> __________________________________
>>> Discover Yahoo!
>>> Have fun online with music videos, cool games, IM
>> and more. Check it
>>> out!
>>> http://discover.yahoo.com/
>>>
>>
> online.html<load_seqdatabase.pl>_______________________________________
>>
>>> ________
>>> Bioperl-l mailing list
>>> Bioperl-l at portal.open-bio.org
>>>
>>
> http://portal.open-bio.org/mailman/listinfo/bioperl-l
>> -- 
>>
> -------------------------------------------------------------
>> Hilmar Lapp                            email: lapp
>> at gnf.org
>> GNF, San Diego, Ca. 92121              phone:
>> +1-858-812-1757
>>
> -------------------------------------------------------------
>>
>>
>>
>
>
>
> 		
> __________________________________
> Discover Yahoo!
> Get on-the-go sports scores, stock quotes, news and more. Check it out!
> http://discover.yahoo.com/mobile.html
>
>
-- 
-------------------------------------------------------------
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