Re: [galaxy-dev] Galaxy cloud: Tool execution generated the following error message
Alicia; Thanks for the useful feedback. I'm agreed that Cloud Galaxy usage requires more IT knowledge compared to running directly on a hosted Galaxy instance. Dannon, Enis and everyone are working hard to make it easier, but at this point it does require understanding Amazon EC2. For your termination problem, you want to be sure to quit using the CloudMan web admin interface. In the terminate dialog, there is a checkbox you can click to delete all of the EBS volumes associated with a cluster. This is off by default to avoid deleting important data, but if you check it this should remove all of your associated data files. For the AMI issue, I have to defer back to Dannon as I'm not sure how they're handling the markupsafe egg conflict with the official AMI. Dannon, happy to coordinate on the new CloudBioLinux if that helps at all. Thanks again for the helpful feedback, Brad
Thank you for your message.
I followed the instructions on "http://wiki.galaxyproject.org/CloudMan". I created an Amazon account, and also ollowed a tutorial video in setting up the account. The wiki page recommended the AMI as below:
*Current AMI:*
- AMI: ami-da58aab3 - Name: 861460482541/galaxy-cloudman-2011-03-22
By the way, I find it quite confusing to use the Amazon Web services (probably due to the fact that I'm not a computer scientist, just a research assistant trying to analyse her data). All the different charges/storage quite are confusing. I'm now being charged for USD10 for a Elastic block storage even though I have deleted all the instances. There is also no way to contact Amazon tech support as you need to pay a subscription fee first. This is very different from Penguin COmputing who works with Life Tech. We are Ion Torrent users, and sometimes use the cloud services, the support at Penguin Comp are very responsive, and there is no hidden costs like the EBS, which are tucked away at another corner, charging me for the storage even though I've deleted the instances. Luckily I keep checking the account statement and managed to find the hidden EBS, and quickly deleted all the volumes that i didn't even know I have created!
I would prefer to use the public server as it seems so much, I think Galaxy is really a very useful tool for non-bioinformaticians like me, and I do not mind having to pay a subscription fee use the public server if funding is a issue. Recently the web-based platform seems to be quite unstable, it hangs or the page just could not be loaded (DNS missing..) That's why i went to use the Galaxy cloud, but it didnt work.
Best regards Alicia
On Fri, Jan 4, 2013 at 9:53 PM, Brad Chapman <chapmanb@50mail.com> wrote:
Alicia;
*I hope this is the correct email to contact re. Galaxy cloud. I tried to run Galaxy on the cloud, I could upload the file but could not go beyond simple function "operate on genomic intervals". Is this normal? This is my first time using the Galaxy on the cloud. Hope to hear from you soon. Thank you. * [...] /mnt/galaxyTools/galaxy-central/lib/galaxy/__init__.py:79: UserWarning: Module markupsafe was already imported from
/usr/local/lib/python2.6/dist-packages/MarkupSafe-0.11-py2.6-linux-x86_64.egg/markupsafe/__init__.pyc,
but /mnt/galaxyTools/galaxy-central/eggs/MarkupSafe-0.12-py2.6-linux-x86_64-ucs4.egg is being added to sys.path
What method did you use to start up the Galaxy cloud instance? I've seen this problem with recent versions of Galaxy and BioCloudCentral, using the current CloudBioLinux AMI (ami-46d4792f). We have a new version of CloudBioLinux/CloudMan in development which fixes this. In the short term, you could try the Galaxy Cloud Launch page to create your instance:
https://main.g2.bx.psu.edu/cloudlaunch
Dannon and Enis, we could isolate Galaxy inside a virtualenv as a more general fix to these egg conflicts.
Hope this helps, Brad
participants (1)
-
Brad Chapman