Anyone any idea why this might happen? It gave this error after being on “Monitoring deployment” for about 5 mins. Github action took 7 mins in total before it failed.
Just experienced this as well: same message and syndrome (multi-minute hang). Deployment apparently did succeed on the fly side.
How did you confirm this? Does fly status show that there’s a new version?
Because for me it still says
ID PROCESS VERSION REGION DESIRED STATUS HEALTH CHECKS RESTARTS CREATED
50cf4ff2 app 23 ams run running 1 total, 1 passing 0 2022-12-18T17:34:26Z
(Old deployment from yesterday)
I just checked my app and definitely there’s not a new version online 
Oh, the plot thickens, sorry, I misspoke: The fly dashboard still shows an old version.
However, our app at runtime is showing the new version (we stamp it with a unique version id as a docker build arg during deploy).
Hm alright, thanks. Not for me Guess I gotta wait for someone from Fly to help out!
Yes, something seems very inconsistent currently on the fly side.
I tried fly ssh console for the app in question, but it fails, reporting Error no instances found for <app>.
But the app is online and responding to my requests, and I can see runner (in logs) relaunching it around the time of the missing deploy…
I just tried again and got “remote builder unavailable”.
After killing the builder, I got this new error:
DEBUG result image:<nil> error:error building: failed commit on ref "layer-sha256:0b5db87f5b42af9f258f14f367616814cb9b518ea0141f46bdd2706bb256d408": "layer-sha256:0b5db87f5b42af9f258f14f367616814cb9b518ea0141f46bdd2706bb256d408" failed size validation: 0 != 51844146: failed precondition
DEBUG --> POST https://api.fly.io/graphql
{
"query": "\nmutation ResolverFinishBuild ($input: FinishBuildInput!) {\n\tfinishBuild(input: $input) {\n\t\tid\n\t\tstatus\n\t\twallclockTimeMs\n\t}\n}\n",
"variables": {
"input": {
"appName": "tech-ster",
"buildId": "486724",
"builderMeta": {
"builderType": "",
"buildkitEnabled": true,
"dockerVersion": "20.10.12",
"platform": "linux/x86_64/3.15.0",
"remoteAppName": "",
"remoteMachineId": ""
},
"clientMutationId": "",
"finalImage": {
"id": "",
"sizeBytes": 0,
"tag": ""
},
"logs": "error building: failed commit on ref \"layer-sha256:0b5db87f5b42af9f258f14f367616814cb9b518ea0141f46bdd2706bb256d408\": \"layer-sha256:0b5db87f5b42af9f258f14f367616814cb9b518ea0141f46bdd2706bb256d408\" failed size validation: 0 != 51844146: failed precondition",
"machineId": "",
"status": "failed",
"strategiesAttempted": [
{
"error": "",
"note": "no buildpack builder configured, skipping",
"result": "failed",
"strategy": "Buildpacks"
},
{
"error": "error building: failed commit on ref \"layer-sha256:0b5db87f5b42af9f258f14f367616814cb9b518ea0141f46bdd2706bb256d408\": \"layer-sha256:0b5db87f5b42af9f258f14f367616814cb9b518ea0141f46bdd2706bb256d408\" failed size validation: 0 != 51844146: failed precondition",
"note": "",
"result": "failed",
"strategy": "Dockerfile"
}
],
"timings": {
"buildAndPushMs": 81219,
"buildMs": 81193,
"builderInitMs": 0,
"contextBuildMs": 3,
"imageBuildMs": 81189,
"pushMs": -1
}
}
},
"operationName": "ResolverFinishBuild"
}
DEBUG {0x14000af01e0}
DEBUG <-- 200 https://api.fly.io/graphql (259.15ms)
{
"data": {
"finishBuild": {
"id": "486724",
"status": "failed",
"wallclockTimeMs": 81510
}
}
}
Error failed to fetch an image or build from source: error building: failed commit on ref "layer-sha256:0b5db87f5b42af9f258f14f367616814cb9b518ea0141f46bdd2706bb256d408": "layer-sha256:0b5db87f5b42af9f258f14f367616814cb9b518ea0141f46bdd2706bb256d408" failed size validation: 0 != 51844146: failed precondition```I restarted the action again and it seems to work now.
ec1e9485 app 27 ⇡ ams run pending 0 14s ago
50cf4ff2 app 23 ams run running 1 total, 1 passing 0 2022-12-18T17:34:26Z
I got a new VM showing. Would love to know what went wrong here though.
We’re troubleshooting some sync issues that are causing stale data for some apps. This is probably what’s happening here.
1 Like
Cool, thanks. I’ll be dumping my observations here in the hope it will be helpful for debugging.
So it kinda proceeded now, but now my new VM is in “pending” for the last 6 mins:
ec1e9485 app 27 ⇡ ams run pending 0 5m39s ago
Yea it’s just hanging now. This is a reminder for everyone to put timeouts on their Github actions
This is happening to my new project on Fly. Was deploying fine 1 hour ago, now it crashes with this message:
Preparing to run: `/app/bin/migrate` as nobody
2022/12/19 19:10:42 listening on [fdaa:1:a26:a7b:9f:6275:cf41:2]:22 (DNS: [fdaa::3]:53)
19:10:45.408 [info] Migrations already up
Starting clean up.
==> Monitoring deployment
Logs: https://fly.io/apps/myapp/monitoring
Error 1 error occurred:
* No deployment available to monitor
Experiencing the same issue. Had a successful deploy about ~8 hours ago now every new PR fails CI/CD with a No deployment available to monitor message.
Same issue over here too. Applications activity tab shows V57 has been released but the monitor tab still shows v54 as active. Builds timeout “No deployment available to monitor”.
I tried the deployment a few times just incase.
Same issue here, and now the instance behind my app just vaporized, the application running on the instance is still responding fine. The versions reported by the app and the instance were way out of sync before this tho.
So now I have 1 VM according to the overview that is in “pending”:
The other working VM is still there though, it just doesn’t show up in the overview anymore
Sorry about all this folks. As @kurt said we we had some issues with v1 apps this afternoon, specifically with syncing which caused stale data. We’ve implemented a fix and are monitoring. status.flyio.net
1 Like
@DAlperin I’m assuming there’s a V2 app type now? I just created my app yesterday and cannot deploy either. Does this mean my Fly app is v1 somehow? I just used fly launch locally.
How do I check if my app is v1 or v2?