Ah, ok, thanks. That’s weird to have it as true by default as it’s unlikely that I want to have this behaviour on a simple hosting. It doesn’t matter how much capacity the machine has — if I am able to write an app that can operate on that capacity then it’s not low — it’s normal.
Updating to flyctl v0.1.172 should unstick deployment for apps without volume storage. Now, if a host doesn’t have capacity to update a Machine (that’s not using a volume) in place on deployment, flyctl will attempt to replace that Machine with a new one on a different host.
@catflydotio / @JP_Phillips for me it is an app without volumes and I do have auto_stop_machine on true. I just tried it again and it failed. Then I upgraded the fly version and now it works. Thanks guys.
@mif In case you want to reset your config try to upgrade also
Hi there,
I have a related issue I guess: my app doesn’t start (after auto_stop) :
2023-08-09T09:27:42.991 proxy[e82d524f716948] cdg [info] Starting machine
2023-08-09T09:27:43.018 proxy[e82d524f716948] cdg [error] machines API returned an error: "could not reserve resource for machine: insufficient memory available to fulfill request"
2023-08-09T09:27:45.024 proxy[4d891090a32968] cdg [info] Starting machine
2023-08-09T09:27:45.050 proxy[4d891090a32968] cdg [error] machines API returned an error: "could not reserve resource for machine: insufficient memory available to fulfill request"
I have been trying to restart it manually by cli but it didn’t work neither :
Error: could not start machine e82d524f716948: failed to start VM e82d524f716948: aborted: could not reserve resource for machine: insufficient memory available to fulfill request
When I’m doing fly machines status, I got this :
C:\Users\Pierre\display_geodata_api>fly machines status e82d524f716948
Machine ID: e82d524f716948
Instance ID: 01H2T6HGDJ5XTHWQ4WRMX4Q2X2
State: failed
VM
ID = e82d524f716948
Instance ID = 01H2T6HGDJ5XTHWQ4WRMX4Q2X2
State = failed
Image = display-geodata-api:deployment-01H2Q4PDJTZF1EA13PW3872QRT
Name = patient-mountain-7731
Private IP = fdaa:0:ef46:a7b:126:174f:4807:2
Region = cdg
Process Group = app
CPU Kind = shared
vCPUs = 4
Memory = 1024
Created = 2023-06-06T15:22:42Z
Updated = 2023-08-09T13:11:51Z
Entrypoint =
Command =
Event Logs
STATE EVENT SOURCE TIMESTAMP INFO
failed launch flyd 2023-08-09T15:11:51.614+02:00
starting start user 2023-08-09T15:11:51.588+02:00
failed launch flyd 2023-08-09T15:09:06.724+02:00
starting start user 2023-08-09T15:09:06.706+02:00
failed launch flyd 2023-08-09T15:03:32.723+02:00
Maybe someone can help ?
Thx!
A fly deploy should get your app back, see Update on `could not reserve resource for machine` error
thx @dangra , as simple as that, it fixed it (I read that post earlier but obviously too fast…)
Seeing this now in nrt region… can’t deploy…
Also seeing this error deploying to LHR location:
“Updating existing machines in ‘[project-name]’ with immediate strategy
([id] [app]) Continuing after error: failed to update VM [vm-id]: aborted: could not reserve resource for machine: insufficient memory available to fulfill requestfailed to update VM [vm-id]”
Edit: Read nina’s post, which is useful knowledge. Still leaves me have to figure out how to get redeployed and building. Unfortunately fly deploy alone doesn’t resolve the issue.