How to restart a previous Cloudman instance

Question: How to restart a previous Cloudman instance

4.1 years ago by

United States

ravi.alla • 10 wrote:

Hi,

I am trying to start a galaxy instance in the cloud using AWS. I managed to follow instructions from https://wiki.galaxyproject.org/CloudMan and get an instance of galaxy going. I've uploaded some data and ran some analysis. I now want to terminate my cluster and cloud instance and log back in at a later time. How do I do this? I shut off my master node (but didn't delete the cluster).

Instructions ask me to use the same cluster name and credentials. If my first cluster name was say XYZ_CLOUD, should I use this same name when I give my AWS secret credentials on Galaxy's launch cloud instance site? and also the same password? I tried this once but my data was not saved. I get a new galaxy and cloud interface address and all my data is lost. What am I doing wrong here?

Thanks

cloudman • 1.0k views

ADD COMMENT • link •

modified 4.1 years ago • written 4.1 years ago by ravi.alla • 10

4.1 years ago by

Dannon Baker ♦ 3.7k

United States

Dannon Baker ♦ 3.7k wrote:

Correct, all you need to do to restart the same cluster is use the same cluster name when launching, so, XYZ_CLOUD. You can use a different password and even instance type if you'd like. The only thing you need to make sure of is that when you shut down the cluster you do *not* check the permanently delete option (unchecked by default).

I'd be surprised if the data was actually lost -- unless those snapshots were manually deleted from the AWS account or something like that (which would produce an error in cloudman upon launch), it's probably still around. When that instance came back up, did you have to re-initialize it? That is, did you get the cloudman screen asking what type and size of cluster you'd like? Or did Galaxy come back up automatically, and it just didn't have your data?

ADD COMMENT • link written 4.1 years ago by Dannon Baker ♦ 3.7k

4.1 years ago by

ravi.alla • 10

United States

ravi.alla • 10 wrote:

Dannon,

Thank you for the answer. I totally missed the dropdown in cloud launcher where it says instances in your account. I forgot to click on that dropdown and pick an already existing galaxy. Instead I kept it as new cluster and used the same name as before. This did ask me to reinitialize the cloud instance (just like the first time). It probably rewrote over the existing cloud instance with the same name. I seem to have fixed this issue when I selected the pre-existing galaxy instance from the dropdown.

Thanks

ADD COMMENT • link modified 4.1 years ago • written 4.1 years ago by ravi.alla • 10

Please log in to add an answer.

Similar posts • Search »