It looks like on July 26th “Fly Admin” made 3 deploys of our app, and then 4 days later on the 30th it went down. I’m not sure why, but it’s in “suspended” mode now. Anyway, we ignored it for 2 weeks because it’s not mission critical and we were doing other things, but I’m trying to get it back up now. I tried scaling to 0 and then scaling back to 1 and I got:
Executing scale plan
Using unattached volume vol_NAME
Error: failed to launch VM: aborted: could not reserve resource for machine: insufficient memory available to fulfill request
Which isn’t great. It sounds like either YYZ is at capacity, or the error is misleading. Thoughts?
It would be really nice if I didn’t have to delete my volume and sync it from scratch like I did 9 months ago when my app died for no reason the last time…
If your app uses a volume, it may have been auto-stopped, and when it attempts to restart, it can’t because the particular host on which it used to live doesn’t have enough capacity for it.
If an app does not use a volume, it can be instantiated on a different host with enough capacity; so the main blocker here is very likely the use of volumes.
There’s a more detailed explanation here complete with some suggestions on what to do, the “Move machines with attached volumes” section would apply. Just beware, if you do have data you need to keep in your volume, you need to clone --from-snapshot to ensure your data is brought over, and ensure the new machine does have all your old data before deleting the old one and its volume.
The clone will instantiate the new machine on a different host with enough capacity.
Ok. When I list the volumes I get a blank “attached to” field, so I assumed the volume wasn’t constrained to anything, but you’re saying it’s linked to a particular machine anyway?
So I’d have to destroy it and clone from a snapshot in order to get it into a new machine, which will then boot?