Hi, Dannon.

        Thank you for your reply. I was using this form:

which was in turn referenced in this tutorial:


        It looks like there is a lot of outdated stuff on the net regarding the 
Galaxy / Cloudman bundle. Anyway, I followed the flow of this new form and got 
my instance up and running in a few minutes (I chose a Galaxy-type cluster). 
However, the Galaxy, Postgres and File System services seemed to be stuck at 
the "Unstarted" status. I tried to manually start them and got the following 
error message at the Galaxy log:

 RuntimeWarning: __builtin__.file size changed, may indicate binary 
  from csamtools import *
python path is: 
/mnt/galaxyTools/galaxy-central/lib, /usr/lib/python2.6/, /u!
 sr/lib/python2.6/plat-linux2, /usr/lib/python2.6/lib-tk, /usr/!
on2.6/lib-old, /usr/lib/python2.6/lib-dynload
Traceback (most recent call last):
  File "/mnt/galaxyTools/galaxy-central/lib/galaxy/webapps/galaxy/buildapp.py", 
line 36, in app_factory
    app = UniverseApplication( global_conf = global_conf, **kwargs )
  File "/mnt/galaxyTools/galaxy-central/lib/galaxy/app.py", line 29, in __init__
  File "/mnt/galaxyTools/galaxy-central/lib/galaxy/config.py", line 315, in 
    raise ConfigurationError( "Unable to create missing directory: %s\n%s" % ( 
path, e ) )
ConfigurationError: Unable to create missing directory: /mnt/galaxyData/files
[Errno 13] Permission denied: '/mnt/galaxyData/files'
Removing PID file paster.pid

        I'm still wondering if I might be doing something wrong, which would 
surprise me as this time the process required much less interaction and input 
than the previous ones I was following.

        Any thoughts?


-----Original Message-----
From: Dannon Baker [mailto:dannonba...@me.com] 
Sent: Friday, December 07, 2012 6:19 AM
To: Fabiano Lucchese
Cc: galaxy-dev@lists.bx.psu.edu
Subject: Re: [galaxy-dev] Installation issue on EC2

Which web form did you use to start the cloud instance?  The page at 
usegalaxy.org/cloudlaunch will always point to the currently supported AMI, and 
I'd definitely recommend using this form for launching galaxy instances.

The problem you're running into is that the unsupported AMI already initialized 
your database as postgres9, which isn't backwards compatible with the postgres8 
installed on the official AMI.  At this point, if you want to continue with 
this specific cluster you'll need to keep using that original AMI (or attempt 
to manually migrate the database between instances, which I wouldn't recommend).

If you don't have any data you care to preserve on this cluster, just delete it 
and start over using usegalaxy.org/cloudlaunch and you should be good to go. If 
you want to try again fresh without deleting anything, just use a new cluster 
name when you launch the instance.


On Dec 6, 2012, at 7:45 PM, Fabiano Lucchese <fabiano.lucch...@hds.com> wrote:

>                 Hi, all.
>                 I've been trying to deploy a CloudMan/Galaxy cluster on EC2 
> for a couple of days now and have been facing all kinds of unexpected 
> weirdness's. I found out that the web-based form that automatically creates 
> the EC2 instance uses an image that doesn`t seem to be officially supported 
> (or at least encouraged) by the Galaxy team, so I decided to launch the 
> instance by hand following the instructions from here:
> http://wiki.galaxyproject.org/CloudMan/AWS/GettingStarted
>                 Up to the point before accessing Galaxy, everything looks ok, 
> but the service never starts. Instead, it shows up at the CloudMan Admin page 
> as "Error" and nothing happens if I try to Stop/Start or Restart it.There's 
> no log either.  The Postgres service displays an immutable "Starting" status, 
> while its log provides the following message:
>                 FATAL:  database files are incompatible with server
> DETAIL:  The data directory was initialized by PostgreSQL version 9.1, which 
> is not compatible with this version 8.4.7.
>                 It does look like a messed up installation, which is quite 
> surprising as this is supposed to be the recommended working image.
>                 Does anybody have any clue on what might be going on?
>                 Thanks in advance.
>                 F.
> ___________________________________________________________
> Please keep all replies on the list by using "reply all"
> in your mail client.  To manage your subscriptions to this and other 
> Galaxy lists, please use the interface at:
>  http://lists.bx.psu.edu/

Please keep all replies on the list by using "reply all"
in your mail client.  To manage your subscriptions to this
and other Galaxy lists, please use the interface at:


Reply via email to