You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When you initiate a deploy it usually runs fine, but sometimes it doesn't even spin up. Usually this is a permissions thing, and the deployment doesn't even really 'start', just sits there endlessly in the 'running' state.
The issue is that when this happens, you can't cancel the deployment. I'm guessing that because the system doesn't register the deployment as having really 'started' the cancel button doesn't show up. But because the deployment fell over, it'll never stop. The end result is that you can never cancel it and it stays in 'running' forever.
What I found then was while I was continually having to re-test (while fixing the permissions thing) , each time had to break the lock - because officially the deployment hadn't 'finished' even though the deploy process was clearly stillborn :(
Obviously this is an aesthetic issue rather than something that actually stops use - and thus not very high priority at all. But I thought you might want to know of what happens in the pathalogical case. :)
Taryn
The text was updated successfully, but these errors were encountered:
insoul
pushed a commit
to insoul/webistrano
that referenced
this issue
Oct 28, 2013
When you initiate a deploy it usually runs fine, but sometimes it doesn't even spin up. Usually this is a permissions thing, and the deployment doesn't even really 'start', just sits there endlessly in the 'running' state.
Sorry for pointing a lot at my blog, but if you need an example of this happening:
http://rubyglasses.blogspot.com/2009/12/getting-webistrano-to-deploy-under.html
The issue is that when this happens, you can't cancel the deployment. I'm guessing that because the system doesn't register the deployment as having really 'started' the cancel button doesn't show up. But because the deployment fell over, it'll never stop. The end result is that you can never cancel it and it stays in 'running' forever.
What I found then was while I was continually having to re-test (while fixing the permissions thing) , each time had to break the lock - because officially the deployment hadn't 'finished' even though the deploy process was clearly stillborn :(
Obviously this is an aesthetic issue rather than something that actually stops use - and thus not very high priority at all. But I thought you might want to know of what happens in the pathalogical case. :)
Taryn
The text was updated successfully, but these errors were encountered: