Error: failed to update VM xxxxxx: request returned non-2xx status, 504 (Request ID: yyyyyy-sin)

1

I’m new to fly.io and it’s the first time I try to deploy my app. fly deploy gives me the error:

Error: failed to update VM xxxxxx: request returned non-2xx status, 504 (Request ID: yyyyyy-sin)

When executing with log verbose:

set “LOG_LEVEL=debug” && fly deploy

[...]
DEBUG <-- 504 https://api.machines.dev/v1/apps/myapp/machines/xxxxxx
[...]
DEBUG <-- 200 https://api.fly.io/graphql (563.07ms)

DEBUG {
  "data": {
    "updateRelease": {
      "release": {
        "id": "zzzzzz"
      }
    }
  }
}

DEBUG Task manager done
Error: failed to update VM xxxxxx: request returned non-2xx status, 504 (Request ID: yyyyyy-sin)

Before it fails it outputs the following information:

→ Pushing image done
image: registry.fly.io/myapp:deployment-xxxxxx
image size: 125 MB

I noticed that the machine is in state “suspended”, but I can’t figure out what I should do about it.

Any help? Thank you

2

Possible related:

3

You have been stuck for 2 week? Are you on a paid plan? Did you get an answer from support?

I guess I misconfigured something but the platform doesn’t give me any information or hint of a potential misconfiguration.

4

No for longer, but only sometimes. Our deploy pipeline just keeps breaking.

We are on the highest plan, aside from enterprise. I’ve emailed support, but they’ve been unable so far to find a root cause. I got some tips like change the registry and it helps sometimes but not always.

5

I have a few suggestions for you though, as a short term fix. Try these:

  • Kill the builder. Just go to the dashboard, then to the builder, and delete it. A new one will be created when deploying.
  • Kill the machine. fly m list and fly m destroy <id> --force
2 Likes
6

Been getting the same Error: failed to update VM xxxxxx: request returned non-2xx status, 504 (Request ID: yyyyyy-sin) error since earlier today (was working fine until last friday), in a v2 app with multiple machines/process groups - one of the machines fail to update; now trying @Yaeger’s suggestion of killing the builder and machine…

7

… and now the specific machine gets stuck at 2023-10-23T18:26:42.627 runner[xxxxxx] sea [info] Pulling container image registry.fly.io/nnnnnnnnnn:deployment-zzzzzzzzzzzzz. The deployment times out, and that line is the last one shown in monitoring.

8

Update: took 34 minutes but it finished deploying :sweat_smile:: 2023-10-23T19:00:50.329 runner[xxxxxx] sea [info] Successfully prepared image registry.fly.io/nnnnnnnnnn:deployment-zzzzzzzzzzzzz (34m7.702432238s)

1 Like
9

Hi, We are also experiencing this quite often but it is intermittent issue on our end.

10

Good to know.

I’m stuck at the “pulling container image” part too.

Will check in 30min :smiley:

11

I’ve been talking to support a lot and got this message recently:

A bit of insight into what’s happening on our end – over the past few days, a significant amount of our engineering focus has shifted to improving reliability for Machine placement and the registry. We have some people focusing 100% of their efforts on this, so I’d expect to see improvements in the coming weeks. I will keep you updated on the progress.

That’s good to hear :slight_smile:

12

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.