It doesn’t look like the fly proxy can start my app, I keep getting:
{
“id”: “0000000000000000acd9f57a11858d6b30a70ca8a9f736a1-1774981010109021000”,
“message”: “[PM01] machines API returned an error: “could not reserve resource for machine: insufficient CPUs available to fulfill request””,
“timestamp”: “2026-03-31T18:16:50.109Z”,
“level”: “error”,
“instance”: “2876d22b464638”,
“provider”: “proxy”,
“region”: “iad”,
“process_group”: null
}
{
“id”: “cfd39f8c-2e36-4979-be52-5a3c0c351977”,
“message”: “[PR03] could not find a good candidate within 1 attempts at load balancing. last error: [PM01] machines API returned an error: “could not reserve resource for machine: governor policy blocked start: hard gate failed: cpu_util_exceeds_target (CPU utilization exceeds target + buffer for existing)””,
“timestamp”: “2026-03-31T18:17:10.374Z”,
“level”: “error”,
“instance”: “2876d22b464638”,
“provider”: “proxy”,
“region”: “iad”,
“process_group”: null
}
It is the smallest size you offer shared-cpu-1x@256MB
Hi… I just tried creating a Machine that size in iad and didn’t encounter any problems. The capacity crunch is probably limited to the one physical host machine that you’re on—not the entirety of iad.
(This is part of why Fly.io recommends always having multiple Machines, incidentally.)
You can either wait for your Machine to be auto-migrated or follow the mitigation tips in the official troubleshooting docs:
For now, (unfortunately) auto-migration doesn’t happen for machines with attached volumes, which is likely also why OP is seeing this error (it is normally hidden away by the auto-migration process). For now the recommended solution to this for machines with volumes is to manually fork a new volume from the old one, which is also covered in the docs (though under a different section).