SSH failing on a single app (operation timed out)

Questions / Help
1

Essentially the same issue as these:

I’ve been trying to SSH into one of my apps, and keep seeing this error:

▲ fly ssh console

Connecting to fdaa:9:a7a:a7b:528:b93b:56c:2... complete
Error: error connecting to SSH server: dial: connect tcp [fdaa:9:a7a:a7b:528:b93b:56c:2]:22: operation timed out

As suggested in the other threads, I tried to clone the machine, but the clone operation is also failing (no volumes attached).

I tried restarting the app, scaling the app size, and scaling the app count (to a different region). But I’m still unable to SSH into the app when attempting to connect to the instance in the other region:

▲ fly ssh console -r ams

Connecting to fdaa:9:a7a:a7b:19b:3a34:6441:2... complete
Error: error connecting to SSH server: dial: connect tcp [fdaa:9:a7a:a7b:19b:3a34:6441:2]:22: operation timed out

Note that I’m still able to SSH into a different app on my personal account (which is also in a different region). But not into this one.

2

Machines list:

 ID             │ NAME                │ STATE   │ CHECKS │ REGION │ ROLE │ IMAGE                          │ IP ADDRESS                     │ VOLUME │ CREATED              │ LAST UPDATED         │ PROCESS GROUP │ SIZE
 8ed934ea746428 │ spring-glitter-9369 │ started │ 1/1    │ ams    │      │ allocator-one:release-cee49aba │ fdaa:9:a7a:a7b:19b:3a34:6441:2 │        │ 2026-10-02T04:10:18Z │ 2026-10-02T04:10:40Z │ app           │ performance-4x:8192MB
 7812d31c691718 │ cold-sunset-9416    │ started │ 1/1    │ fra    │      │ allocator-one:release-cee49aba │ fdaa:9:a7a:a7b:528:b93b:56c:2  │        │ 2026-08-15T17:31:06Z │ 2026-10-02T04:08:57Z │ app           │ performance-4x:8192MB
3

I just killed both machines (by scaling count to 0 in both regions). After that, I scaled the app back to 1 (in only one region):

 ID             │ NAME           │ STATE   │ CHECKS │ REGION │ ROLE │ IMAGE                          │ IP ADDRESS                     │ VOLUME │ CREATED              │ LAST UPDATED         │ PROCESS GROUP │ SIZE
 89136ec6433508 │ old-voice-5397 │ created │ 0/1    │ fra    │      │ allocator-one:release-cee49aba │ fdaa:9:a7a:a7b:1c4:e7a1:80d5:2 │        │ 2026-10-02T04:32:33Z │ 2026-10-02T04:32:33Z │ app           │ performance-4x:8192MB

And SSH is still timing out:

▲ fly ssh console

Connecting to fdaa:9:a7a:a7b:1c4:e7a1:80d5:2... complete
Error: error connecting to SSH server: dial: connect tcp [fdaa:9:a7a:a7b:1c4:e7a1:80d5:2]:22: operation timed out

Note that I briefly also saw this error, but haven’t seen it again:

Connecting to tunnel 🌎
Error: tunnel unavailable: Error contacting Fly.io API when probing "allocator-one": timed out (context deadline exceeded)
4

can you run fly doctor and post the output?

5

All looks good with fly doctor:

▲ fly doctor
Testing authentication token... PASSED
Testing flyctl agent... PASSED
Testing local Docker instance... Nope
Pinging WireGuard gateway (give us a sec)... PASSED
Testing WireGuard DNS... PASSED
Testing WireGuard Flaps... PASSED

Everything passes with the -a app flag as well.

6

Some other things I’ve tried so far:

  • fly agent ping passes successfully
  • fly agent restart did not resolve the issue
  • Tried scaling the app in different regions (ams, fra, ord) and tried connecting to region-specfic instances. I still get the same error.
  • Interestingly, my colleague (in Austria) is able to SSH into the app.
  • While I cannot SSH into this specific app, I can SSH into other ones.
  • I’ve sent Fly support the output of these commands (just waiting to hear back):
    • LOG_LEVEL=debug fly ssh console --debug -a <app>
    • curl https://debug.fly.dev-H "flyio-debug: doit" -v
7

Heard back from support. Running the following worked for me:

fly wg reset <app>

After resetting the wg tunnel, I was able to SSH into the app.