- galaxy-dev - lists.galaxyproject.org

Installation of 20130603 Galaxy image with MySQL fails
by Linda Cham 21 Jun '13

21 Jun '13

Hi. I am new to the Galaxy forum and I am trying to setup a new local Galaxy instance with MySQL on RH6.2 x86_64. On a newly installed OS node, using the hg clone command (hg clone https://bitbucket.org/galaxy/galaxy-dist#stable) which gets the current Galaxy 20130603 image, I am getting the following errors when running run.sh: File "build/bdist.linux-x86_64-ucs4/egg/MySQLdb/cursors.py", line 173, in execute self.errorhandler(self, exc, value) File "build/bdist.linux-x86_64-ucs4/egg/MySQLdb/connections.py", line 36, in defaulterrorhandler raise errorclass, errorvalue OperationalError: (OperationalError) (1170, "BLOB/TEXT column 'slug' used in key specification without a key length") 'CREATE INDEX ix_history_slug ON history (slug)' () Reading the forum, someone stated it can be bypass by running the CREATE INDEX. mysql> CREATE INDEX ix_history_slug ON history (slug(1000)); Query OK, 0 rows affected (0.09 sec) Records: 0 Duplicates: 0 Warnings: 0 When trying to run.sh again after the CREATE INDEX, will get this error: File "/gpfs/gpfs2/home/galaxyadmin25/galaxy-dist/lib/galaxy/app.py", line 52, in __init__ create_or_verify_database( db_url, kwargs.get( 'global_conf', {} ).get ( '__file__', None ), self.config.database_engine_options, app=self ) File "/gpfs/gpfs2/home/galaxyadmin25/galaxy-dist/lib/galaxy/model/migrate/check.py", line 107, in create_or_verify_database % ( db_schema.version, migrate_repository.versions.latest, config_arg ) ) Exception: Your database has version '29' but this code expects version '115'. Please backup your database and then migrate the schema by running 'sh manage_db.sh upgrade'. So try to run the "sh manage_db.sh upgrade' now will get this error: $ sh manage_db.sh upgrade Traceback (most recent call last): File "./scripts/manage_db.py", line 13, in <module> from migrate.versioning.shell import main File "/gpfs/gpfs2/home/galaxyadmin25/galaxy-dist/eggs/sqlalchemy_migrate-0.7.2-py2.6.egg/migrate/versioning/shell.py", line 12, in <module> from migrate.versioning import api File "/gpfs/gpfs2/home/galaxyadmin25/galaxy-dist/eggs/sqlalchemy_migrate-0.7.2-py2.6.egg/migrate/versioning/api.py", line 33, in <module> from migrate.versioning import (repository, schema, version, File "/gpfs/gpfs2/home/galaxyadmin25/galaxy-dist/eggs/sqlalchemy_migrate-0.7.2-py2.6.egg/migrate/versioning/repository.py", line 13, in <module> from migrate.versioning import version, pathed, cfgparse File "/gpfs/gpfs2/home/galaxyadmin25/galaxy-dist/eggs/sqlalchemy_migrate-0.7.2-py2.6.egg/migrate/versioning/version.py", line 10, in <module> from migrate.versioning import pathed, script File "/gpfs/gpfs2/home/galaxyadmin25/galaxy-dist/eggs/sqlalchemy_migrate-0.7.2-py2.6.egg/migrate/versioning/pathed.py", line 11, in <module> from migrate.versioning.util import KeyedInstance File "/gpfs/gpfs2/home/galaxyadmin25/galaxy-dist/eggs/sqlalchemy_migrate-0.7.2-py2.6.egg/migrate/versioning/util/__init__.py", line 7, in <module> from decorator import decorator ImportError: No module named decorator In some searches, there were tips of deleting the eggs directory and rerun run.sh or use a python env. Neither option worked. Was not able to find anything meaningful about the decorator error. Note:, I was able to get Galaxy up with MySQL using the Galaxy 788cd3d06541 image on a different node. So I am thinking maybe it is the new 20130603 image that is giving me issues? Commands used to get Galaxy up: Install mysql from yum repository [MySQL 5.1.52-1.el6_0.1] python version [2.6.6] Grant ALL PRIVILEGES galaxy_db.* TO galaxy_user on localhostname with PWD Images tested: *Copy of the galaxy-galaxy-dist-788cd3d06541.tar.bz2 image Did not run the hg update command. [This worked] *hg clone stable of newest image [Failed] Update the universe_wsg.ini with the port, host, database_connection to mysql Ran run.sh Thanks. Linda

4 9

nglims roles
by Lee Katz 21 Jun '13

21 Jun '13

Hi, what roles are embedded in nglims? (am I calling it by the right name?) I have a "sequencing" role which will be given to anyone submitting a request, and that seems to be working just fine. I am also using the yaml form to edit the submission form, and I think it's beautifully implemented. However, I cannot figure out how someone on our MiSeq team will access the submissions unless they are admins of Galaxy. I would like to avoid this situation because the MiSeq team should not have the same permissions that allows them to view all data or install new tools. Is there a role that I can give someone who is actually performing the sequencing? And what would it be? -- Lee Katz, Ph.D.

2 4

visulaization tool
by shashi shekhar 21 Jun '13

21 Jun '13

Hi, i want to add new visualization tool in galaxy . can i get the required documents ? can u send me the details how can i add the visualization tool in galaxy . Regards shashi

5 15

Initial display with Simulate (version 1.0.0)
by Perez, Ricardo 21 Jun '13

21 Jun '13

Dear all, When running Simulate (version 1.0.0)I am missing a graph in the initial display. Is this a bug or is it my instance problem. On a side note, when running Simulate I found out that if you don't place parameters in "Frequency/ies for minor allele:" and "Detection thresholds:" the job will fail. It would be nice to indicate that those parameters are a must for running the tool. Thank you, --Ricardo Perez

1 0

simplest way to connect RNAseq tools for a local Galaxy instance
by Elwood Linney 20 Jun '13

20 Jun '13

I have: 1) downloaded the Galaxy framework into a Mac Pro desktop with 64gb ram 2) downloaded Mac OS versions of Tophat2, Bowtie2, and the Cufflink suite to the mac 3) have looked at the Galaxy wiki on connecting these [http://wiki.galaxyproject.org/Admin/Config/Tool%20Dependencies] 4) I have gotten some useful advice from Dannon Baker on how to introduce my large datasets on the some computer into Galaxy through Admin changing some permissions but my minimal knowledge of unix commands and tricks has me stymied Question: If my primary purpose is to use this local instance for the analysis of RNAseq data within my lab, what is the simplest way of connecting the programs to the Galaxy framework? While from reviewing info on the Galaxy website I am only guessing that adding them to $PATH might be the easiest way for me to do this (but I am not exactly sure what this means or specifically how I would do it in my case). However, I do initiate the running of Galaxy by sh run.sh on my terminal and visualizing things via my localhost port. There is some indication from the wiki I found that this might not be good for a system initiated by sh run.sh . [http://wiki.galaxyproject.org/Admin/Config/Tool%20Dependencies] under "Local Jobs" I would appreciate any specific advice on this. Sincerely, Elwood Linney DUMC

3 2

API updating meta data for a history
by Philip Mabon 20 Jun '13

20 Jun '13

Hey, I was curious if it was possible to allow the API to change tags associated with a history? At the moment in the galaxy code it's forbidden. Having the ability to add tags using the API would be helpful of associating multiple histories to a particular project. So later, it would be easier to 'select' all the histories to perform a bulk action. Ie download all contigs.fasta from histories that tag with 'project_Ecoli_June20_2013'. Thanks Phil

2 1

retrieving file info from '.dat' dataset
by vijayalakshmi 20 Jun '13

20 Jun '13

Hi all, I have written a java tool to which I have to pass an existing file's name. In the XML wrapper, I am choosing the file from the history using <param type="data">. Even though the tool is receiving the filename as an input, it is looking for a ".dat" file from \database\000\<some dataset> How do I retrieve the file I need from the dataset? Thanks, VJ.

2 1

Re: [galaxy-dev] simplest way to connect RNAseq tools for a local Galaxy instance
by Nicholas Kline 20 Jun '13

20 Jun '13

Hi Elwood, I am in a similar situation. Our lab is trying to connect Bowtie version 1 to our local Galaxy framework. We are slowly making progress with the help of people at seqanswers.com You are welcome to review the thread I started at http://seqanswers.com/forums/showthread.php?t=31047 since it might contain information that relates to your situation. If you do not receive help from this mailing list, I recommend posting your question at seqanswers.com Good luck! > Message: 22 > Date: Thu, 20 Jun 2013 11:45:20 -0400 > From: Elwood Linney <ellinney(a)gmail.com> > To: galaxy-dev(a)lists.bx.psu.edu > Subject: [galaxy-dev] simplest way to connect RNAseq tools for a local > Galaxy instance > Message-ID: > <CAC-Zp0yDhx5Dr0ffdtRB-h=uLZR5UZaBkyVh+8riFYemNmwVnw(a)mail.gmail.com> > Content-Type: text/plain; charset="iso-8859-1" > > I have: > 1) downloaded the Galaxy framework into a Mac Pro desktop with 64gb ram > 2) downloaded Mac OS versions of Tophat2, Bowtie2, and the Cufflink suite > to the mac > 3) have looked at the Galaxy wiki on connecting these > [http://wiki.galaxyproject.org/Admin/Config/Tool%20Dependencies] > 4) I have gotten some useful advice from Dannon Baker on how to introduce > my large datasets on the some computer into Galaxy through Admin changing > some permissions > > but my minimal knowledge of unix commands and tricks has me stymied > > Question: > If my primary purpose is to use this local instance for the analysis of > RNAseq data within my lab, what is the simplest way of connecting the > programs to the Galaxy framework? > > While from reviewing info on the Galaxy website I am only guessing that > adding them to $PATH might be the easiest way for me to do this (but I am > not exactly sure what this means or specifically how I would do it in my > case). However, I do initiate the running of Galaxy by sh run.sh on my > terminal and visualizing things via my localhost port. There is some > indication from the wiki I found that this might not be good for a system > initiated by sh run.sh . > > [http://wiki.galaxyproject.org/Admin/Config/Tool%20Dependencies] under > "Local Jobs" > > I would appreciate any specific advice on this. > > Sincerely, > > Elwood Linney > DUMC >

1 0

Default history
by Jordi Vaquero 20 Jun '13

20 Jun '13

Hello everyone, I am managing an instance of galaxy. We want to provide a set of example datasets for being used, we want this is by default datasets accessible for the anonymous user. there is any way to do that? I have tried to work the shared and publish option of history but they don't work as I wanted. Thank you in advanced Jordi

3 2

Developing Tools
by Saliya Ekanayake 20 Jun '13

20 Jun '13

Hi, I am new to development with Galaxy and wonder if someone could kindly help me out with the following. 1.) A step-by-step guide on how to develop a tool 2.) Are tools limited to simple functions that run on a single node? 3.) When Galaxy is run on cloud (with EC2) how does it utilize worker nodes? Is it such that steps in my workflow get load balanced into nodes? Thank you, Saliya -- Saliya Ekanayake esaliya(a)gmail.com Cell 812-391-4914 Home 812-961-6383 http://saliya.org

2 2