Enis,
running the default EC2 AMI (2011-03-22), cluster was started for the first time about ten days ago so this is odd. I'll read up on how to manually update CM.
Thanks!
Oliver
Hi Oliver,
A message in the masthead/top of the page main CloudMan console page
offering to update CloudMan should show up automatically on page refresh
when a version of CloudMan newer than the one currently running one
exists in the default cloudman bucket. This functionality was added a
few months ago now so if your instance is older, it won't work like
that. In that case, it should be fine to manually download cm.tar.gz
from 'cloudman' bucket and upload it to your cluster's bucket (replacing
the current file there - although make a copy of the current file in
case things don't work out as expected) and restart the instance.
Enis
On Tue, Jun 28, 2011 at 4:28 PM, Oliver Hofmann<ohofmann@hsph.harvard.edu <mailto:ohofmann@hsph.harvard.edu>> wrote:
Enis,
> The admin interface is something that I added just last week.
It's not related to an AMI but to Cloudman itself. When was it you
created the given cluster? If it's been in the last 5 days, the link
should be there (take a look at the most recent newsbrief - there is
a snapshot of the admin link & interface there:
https://bitbucket.org/galaxy/galaxy-central/wiki/Features/DevNewsBrief/2011_06_23).
If your cluster was older than that, Cloudman should have a option
in the masthead to update it automatically. If you are to update,
the link should show up. Let me know if neither is the case and
we'll work on it more.
is there a way to update CloudMan through the older admin page as
well (/cloud/admin)? It already has the Update Galaxy button and
passes on the galaxy-central bitbucket URL to an 'update_galaxy'
script, but doesn't seem to trigger anything beyond that (whether
Galaxy itself is running or not).
Best, Oliver
>
> Glad you got the configuration to work,
> Enis
>
> On Tue, Jun 28, 2011 at 3:45 PM, Joseph Hargitai
<joseph.hargitai@einstein.yu.edu<mailto:joseph.hargitai@einstein.yu.edu>> wrote:> From: Enis Afgan [eafgan@emory.edu <mailto:eafgan@emory.edu>] <mailto:galaxy-user@lists.bx.psu.edu>
> Yeap,
>
> grabbed a new one from the tar file in the bucket and allowed Mac
formatting to prevail. Unicode was the culprit.
>
>
> I am missing however the Cloudman Admin tab you are referring to.
Is this a version issue?
> we use: galaxy-cloudman-2011-03-22 (ami-da58aab3)
>
> I do have the Admin tab in galaxy itself, but I do not think that
is what you are referring to.
>
> joe
>
>
> Subject: Re: [galaxy-user] cloud instance drops settings on reboot
>
> It seems that the formatting of the file was changed and Galaxy
is unable to parse the file. The editor you used to edit the file
might have done this once the file was saved. I'd suggest using vi
to check the file and make sure it follows the formatting of
universe_wsgi.ini.sample for example. Then, try uploading the file
again. BTW, it's required for the file to have the .cloud extension.
>
> If all you're trying to do is add Galaxy admin users, the
Cloudman Admin interface I mentioned earlier offers this
functionality and it handles updating of the universe file. If you
want to revert to what you had before, feel free to pull in the
default universe_wsgi.ini.cloud file from 'cloudman' bucket on S3
and upload it to your cluster's bucket.
>
> Enis
>
> On Tue, Jun 28, 2011 at 3:21 PM, Joseph Hargitai
<joseph.hargitai@einstein.yu.edu> From: Enis Afgan [eafgan@emory.edu <mailto:eafgan@emory.edu>] <mailto:galaxy-user@lists.bx.psu.edu><mailto:joseph.hargitai@einstein.yu.edu>> wrote:
> Enis,
>
> thanks! I did edit the file in the bucket as much as adding an
admin user. I uploaded the file back into the bucket but now the
galaxy part is not starting. It is throwing an error:
>
> • 19:01:24 - Setting up Galaxy application
> • 19:01:24 - Retrieved file 'universe_wsgi.ini.cloud' from
bucket 'cm-99e0c68a2fe11e32a592e36abd1f2c38' to
'/mnt/galaxyTools/galaxy-central/universe_wsgi.ini'.
> • 19:01:25 - Retrieved file 'tool_conf.xml.cloud' from
bucket 'cm-99e0c68a2fe11e32a592e36abd1f2c38' to
'/mnt/galaxyTools/galaxy-central/tool_conf.xml'.
> • 19:01:25 - Retrieved file
'tool_data_table_conf.xml.cloud' from bucket
'cm-99e0c68a2fe11e32a592e36abd1f2c38' to
'/mnt/galaxyTools/galaxy-central/tool_data_table_conf.xml.cloud'.
> • 19:01:25 - Starting Galaxy...
> • 19:01:25 - Error invoking Galaxy, running command
'/bin/su - galaxy -c "export SGE_ROOT=/opt/sge; sh
$GALAXY_HOME/run.sh --daemon"' returned code '1' and following
stderr: 'Traceback (most recent call last): File
"./scripts/check_eggs.py", line 22, in c = Crate() File
"/mnt/galaxyTools/galaxy-central/lib/galaxy/eggs/__init__.py", line
259, in __init__ self.galaxy_config = GalaxyConfig() File
"/mnt/galaxyTools/galaxy-central/lib/galaxy/eggs/__init__.py", line
356, in __init__ if self.config.read( GalaxyConfig.config_file ) ==
[]: File "/usr/lib/python2.6/ConfigParser.py", line 286, in read
self._read(fp, filename) File "/usr/lib/python2.6/ConfigParser.py",
line 482, in _read raise MissingSectionHeaderError(fpname, lineno,
line) ConfigParser.MissingSectionHeaderError: File contains no
section headers. file:
/mnt/galaxyTools/galaxy-central/universe_wsgi.ini, line: 1
'\xff\xfe#\x00\n' Traceback (most recent call last): File
"./scripts/fetch_eggs.py", line 27, in c = Crate() File
"/mnt/galaxyTools/galaxy-central/lib/galaxy/eggs/__init__.py", line
259, in __init__ self.galaxy_config = GalaxyConfig() File
"/mnt/galaxyTools/galaxy-central/lib/galaxy/eggs/__init__.py", line
356, in __init__ if self.config.read( GalaxyConfig.config_file ) ==
[]: File "/usr/lib/python2.6/ConfigParser.py", line 286, in read
self._read(fp, filename) File "/usr/lib/python2.6/ConfigParser.py",
line 482, in _read raise MissingSectionHeaderError(fpname, lineno,
line) ConfigParser.MissingSectionHeaderError: File contains no
section headers. file:
/mnt/galaxyTools/galaxy-central/universe_wsgi.ini, line: 1
'\xff\xfe#\x00\n' '
>
> Is the edited bucket file causing an issue? How do I get a new
one or fix this? I did notice that the first uploaded try omitted
the .cloud suffix, but i uploaded it again with the correct suffix.
>
>
> best,
> joe
>
>
>
>
>
>
>
>
> Subject: Re: [galaxy-user] cloud instance drops settings on reboot
>
> Hi Joe,
> This is by design. In order to change the settings in
universe_wsgi.ini, you should edit file universe_wsgi.ini.cloud in
the cluster's bucket on S3. This file get pulled in by Cloudman each
time an instance starts.
>
> As far as restarting Cloudman, there is a (new) Admin panel (link
at top right of the Cloudman console page) that gives you an option
to restart Cloudman (as well as individual application-level services).
>
> Let us know if you have any more questions,
> Enis
>
> On Tue, Jun 28, 2011 at 11:29 AM, Joseph Hargitai
<joseph.hargitai@einstein.yu.edu<mailto:joseph.hargitai@einstein.yu.edu>> wrote:> at usegalaxy.org <http://usegalaxy.org>. Please keep all replies
>
> Hi,
>
> setting the admin user in
> /mnt/galaxyTools/galaxy-central/universe_wsgi.ini
>
> did work after a simple restart of the galaxy service as user
galaxy - ./runs.sh
>
> However, when I restarted the cloud instance (rebooted from
Amazon console) the setting reverted to original.
>
> Is there perhaps another ini file that overrides this one on a
reboot?
>
> Also, is there a proper/other way to restart the service?
>
> best,
> joe
>
>
> ___________________________________________________________
> The Galaxy User list should be used for the discussion of
> Galaxy analysis and other features on the public server> at usegalaxy.org <http://usegalaxy.org>. Please keep all replies
on the list by
> using "reply all" in your mail client. For discussion of
> local Galaxy instances and the Galaxy source code, please
> use the Galaxy Development list:
>
> http://lists.bx.psu.edu/listinfo/galaxy-dev
>
> To manage your subscriptions to this and other Galaxy lists,
> please use the interface at:
>
> http://lists.bx.psu.edu/
>
>
>
> ___________________________________________________________
> The Galaxy User list should be used for the discussion of
> Galaxy analysis and other features on the public server<tel:%2B1%20%28617%29%20365%200984>
on the list by
> using "reply all" in your mail client. For discussion of
> local Galaxy instances and the Galaxy source code, please
> use the Galaxy Development list:
>
> http://lists.bx.psu.edu/listinfo/galaxy-dev
>
> To manage your subscriptions to this and other Galaxy lists,
> please use the interface at:
>
> http://lists.bx.psu.edu/
--
Research Scientist Department of Biostatistics
Associate Director Bioinformatics Core
Harvard School of Public Health
Skype: ohofmann Phone: +1 (617) 365 0984
--Research Scientist Department of Biostatistics
Associate Director Bioinformatics Core
Harvard School of Public Health
Skype: ohofmann Phone: +1 (617) 365 0984