[Biojava-dev] build problem

Mon Aug 9 22:48:18 UTC 2010

Seems your times are comparable with the build machine. The files are
downloaded and store in local temporary directories (as provided by
System.getProperty("java.io.tmpdir") ) If that tmp directory changes,
the files will have to be re-loaded again, otherwise they will be
re-used and the code runs quicker. Seems the VM is changing the tmp
dir location frequently. Anybody has a suggestion how to define a more
"stable" tmp dir locations? Otherwise I will put all required test
files into the test resources dir ...

Andreas

On Mon, Aug 9, 2010 at 3:20 PM, George Waldon <gwaldon at geneinfinity.org> wrote:
> Here is my summary and some of the timings are pretty good (simple HP laptop with a dual core):
>
> ------------------------------------------------------------------------
> Reactor Summary:
> ------------------------------------------------------------------------
> biojava ............................................... SUCCESS [3.820s]
> bytecode .............................................. SUCCESS [6.022s]
> core .................................................. SUCCESS [52.822s]
> alignment ............................................. SUCCESS [4.332s]
> blast ................................................. SUCCESS [27.453s]
> biojava3-structure .................................... SUCCESS [7:33.228s]
> das ................................................... SUCCESS [18.456s]
> sequence .............................................. SUCCESS [0.122s]
> sequence-core ......................................... SUCCESS [3.974s]
> sequence-rna .......................................... SUCCESS [0.313s]
> sequence-biosql ....................................... SUCCESS [2.735s]
> sequence-fasta ........................................ SUCCESS [0.324s]
> sequence-blastxml ..................................... SUCCESS [3.339s]
> sequencing ............................................ SUCCESS [5.637s]
> phylo ................................................. SUCCESS [5.084s]
> biosql ................................................ SUCCESS [6.344s]
> gui ................................................... SUCCESS [6.552s]
> biojava3-core ......................................... SUCCESS [8.477s]
> biojava3-phylo ........................................ SUCCESS [3.596s]
> biojava3-structure-gui ................................ SUCCESS [8.283s]
> biojava3-alignment .................................... SUCCESS [6.002s]
> biojava3-genome ....................................... SUCCESS [3.933s]
> biojava3-protmod ...................................... SUCCESS [8:59.048s]
> biojava3-ws ........................................... SUCCESS [1.367s]
> ------------------------------------------------------------------------
> ------------------------------------------------------------------------
> BUILD SUCCESSFUL
> ------------------------------------------------------------------------
> Total time: 19 minutes 33 seconds
> Finished at: Mon Aug 09 13:48:00 PDT 2010
> Final Memory: 116M/363M
> ------------------------------------------------------------------------
>
> The long time comes apparently from fetching all these files. I tried the build last week on a different network after removing the faulty test as suggested by Scott and I had similar timing. This could be an issue with NetBeans in fact. If someone has experienced such long delays, it would be interesting to know in which conditions this occurred.
>
> Thanks again,
>
> George
>
>>----- ------- Original Message ------- -----
>>From: Andreas Prlic <andreas at sdsc.edu>
>>To: George Waldon <gwaldon at geneinfinity.org>
>>Sent: Mon, 9 Aug 2010 14:27:27
>>
>>That seems like a very long time... on the
>>automated build server the
>>times look like below:
>>
>>If the structure module takes so much time, I
>>wonder if you are behind
>>a very slow network? Some of the junit tests fetch
>>PDB files from a
>>public ftp server and I wonder if there is a
>>networking issue. Having
>>said this, the two slowest modules are structure
>>and protmod. We will
>>try to cut down the time spent on those tests...
>>
>>Andreas
>>
>>
>>INFO] biojava
>>...............................................
>>SUCCESS [21.430s]
>>[INFO] bytecode
>>..............................................
>>SUCCESS [44.778s]
>>[INFO] core
>>..................................................
>>SUCCESS
>>[4:54.691s]
>>[INFO] alignment
>>.............................................
>>SUCCESS [48.048s]
>>[INFO] blast
>>.................................................
>>SUCCESS
>>[1:20.263s]
>>[INFO] biojava3-structure
>>.................................... SUCCESS
>>[8:12.412s]
>>[INFO] das
>>...................................................
>>SUCCESS
>>[1:13.103s]
>>[INFO] sequence
>>..............................................
>>SUCCESS [18.176s]
>>[INFO] sequence-core
>>......................................... SUCCESS
>>[47.424s]
>>[INFO] sequence-rna
>>.......................................... SUCCESS
>>[28.770s]
>>[INFO] sequence-biosql
>>....................................... SUCCESS
>>[47.032s]
>>[INFO] sequence-fasta
>>........................................ SUCCESS
>>[29.075s]
>>[INFO] sequence-blastxml
>>..................................... SUCCESS
>>[44.392s]
>>[INFO] sequencing
>>............................................
>>SUCCESS [49.520s]
>>[INFO] phylo
>>.................................................
>>SUCCESS [48.459s]
>>[INFO] biosql
>>................................................
>>SUCCESS [57.503s]
>>[INFO] gui
>>...................................................
>>SUCCESS [58.610s]
>>[INFO] biojava3-core
>>......................................... SUCCESS
>>[1:05.398s]
>>[INFO] biojava3-phylo
>>........................................ SUCCESS
>>[46.623s]
>>[INFO] biojava3-structure-gui
>>................................ SUCCESS
>>[1:06.737s]
>>[INFO] biojava3-alignment
>>.................................... SUCCESS
>>[54.682s]
>>[INFO] biojava3-genome
>>....................................... SUCCESS
>>[49.844s]
>>[INFO] biojava3-protmod
>>...................................... SUCCESS
>>[10:05.171s]
>>[INFO] biojava3-ws
>>........................................... SUCCESS
>>[44.379s]
>>
>>
>>On Mon, Aug 9, 2010 at 1:52 PM, George Waldon
>><gwaldon at geneinfinity.org> wrote:
>>> Thanks to all for the fixing.
>>>
>>> The build took 19 minutes and 33 seconds, of
>>which 16 min and 32 s were for the structure
>>modules! This sounds a bit long to me. Is-this
>>expected?
>>>
>>> George
>>>
>>>>----- ------- Original Message ------- -----
>>>>From: Andreas Prlic <andreas at sdsc.edu>
>>>>To: George Waldon <gwaldon at geneinfinity.org>
>>>>Sent: Mon, 9 Aug 2010 12:35:56
>>>>
>>>>If you update the class, this should be fixed
>>now.
>>>>Was a problem with
>>>>a hard coded /tmp path...
>>>>
>>>>Andreas
>>>>
>>>>
>>>>
>>>>On Mon, Aug 9, 2010 at 12:26 PM, George Waldon
>>>><gwaldon at geneinfinity.org> wrote:
>>>>> Hi,
>>>>>
>>>>> I am getting the following failures in
>>>>org.biojava.bio.structure.align.benchmark.Multipl
>>eA
>>>>lignmentTest:
>>>>>
>>>>> Fetching
>>>>ftp://ftp.wwpdb.org/pub/pdb/data/structures/all/p
>>db
>>>>/pdb1hcy.ent.gz
>>>>> writing to \tmp\hc\pdb1hcy.ent.gz
>>>>> java.io.FileNotFoundException:
>>>>\tmp\hc\pdb1hcy.ent.gz (The system cannot find
>>the
>>>>path specified)
>>>>> ...
>>>>>
>>>>> and
>>>>>
>>>>> Fetching
>>>>ftp://ftp.wwpdb.org/pub/pdb/data/structures/all/p
>>db
>>>>/pdb1nls.ent.gz
>>>>> writing to \tmp\nl\pdb1nls.ent.gz
>>>>> java.io.FileNotFoundException:
>>>>\tmp\nl\pdb1nls.ent.gz (The system cannot find
>>the
>>>>path specified)
>>>>> at java.io.FileOutputStream.open(Native Method)
>>
>>>>> at
>>>>java.io.FileOutputStream.<init>(FileOutputStream.
>>ja
>>>>va:179)
>>>>> at
>>>>java.io.FileOutputStream.<init>(FileOutputStream.
>>ja
>>>>va:131)
>>>>> at
>>>>org.biojava.bio.structure.io.PDBFileReader.downlo
>>ad
>>>>PDB(PDBFileReader.java:430)
>>>>> ...
>>>>>
>>>>> Does anyone has a solution for this? I am
>>>>building from within NetBeans.
>>>>>
>>>>> Thanks,
>>>>> George
>>>>>
>>>>
>>>>
>>>>
>>>>--
>>>>-------------------------------------------------
>>--
>>>>--------------------
>>>>Dr. Andreas Prlic
>>>>Senior Scientist, RCSB PDB Protein Data Bank
>>>>University of California, San Diego
>>>>(+1) 858.246.0526
>>>>-------------------------------------------------
>>--
>>>>--------------------
>>>
>>
>>
>>
>>--
>>---------------------------------------------------
>>--------------------
>>Dr. Andreas Prlic
>>Senior Scientist, RCSB PDB Protein Data Bank
>>University of California, San Diego
>>(+1) 858.246.0526
>>---------------------------------------------------
>>--------------------
>

-- 
-----------------------------------------------------------------------
Dr. Andreas Prlic
Senior Scientist, RCSB PDB Protein Data Bank
University of California, San Diego
(+1) 858.246.0526
-----------------------------------------------------------------------