# Can't restart or access machine

**URL:** <https://community.fly.io/t/cant-restart-or-access-machine/27863>\
**Category:** Questions / Help\
**Tags:** postgres, machines\
**Created:** [May 13, 2026, 12:54pm UTC](https://community.fly.io/t/cant-restart-or-access-machine/27863 "2026-05-13T12:54:30Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![c00kie17](https://sea1.discourse-cdn.com/fly1/user_avatar/community.fly.io/c00kie17/32/46826_2.png) [@c00kie17](https://community.fly.io/u/c00kie17)\
**Post date:** [May 13, 2026, 12:54pm UTC](https://community.fly.io/t/cant-restart-or-access-machine/27863/1 "2026-05-13T12:54:30Z")

</div>

My Postgres machine has been unreachable for ~2 hours and I cannot recover it through any CLI commands.

**App:** `guido-postgres`  
**Machine:** `d8939ddb600e48`  
**Region:** `nrt`

**Symptoms:**

- `host_status: "unreachable"` in machines API

- `config: null` returned for the machine

- `fly machine restart` → 408 timeout

- `fly machine stop` → 408 timeout

- `fly ssh console` → flyctl nil pointer panic (because config is null)

- Manual restart via dashboard → “Lease Failed”

- Releases v14 and v15 stuck in `running` state since 12:40 UTC today

**Last request ID:** `01KRGNFGST33HYMYH3R3H97352-nrt`

My backend app `guido-backend` is down because it can’t reach Postgres. Any help appreciated.

---

<div class="post-metadata">

**Author:** ![roadmr](https://sea1.discourse-cdn.com/fly1/user_avatar/community.fly.io/roadmr/32/14140_2.png) [@roadmr](https://community.fly.io/u/roadmr)\
**Post date:** [May 13, 2026, 2:45pm UTC](https://community.fly.io/t/cant-restart-or-access-machine/27863/3 "2026-05-13T14:45:49Z")

</div>

Hi! This host seems to be unresponsive, we’re looking into it. You should get a notification in your [fly.io](http://fly.io/) dashboard with more details if it takes us longer than a few minutes to bring the host back up.

This kind of thing is precisely why we recommend multiple-machine Postgres setups for critical data! If your Postgres server going down causes problems, it means you do care about uptime, and a multi-machine setup will be more resilient to host failures.

---

<div class="post-metadata">

**Author:** ![c00kie17](https://sea1.discourse-cdn.com/fly1/user_avatar/community.fly.io/c00kie17/32/46826_2.png) [@c00kie17](https://community.fly.io/u/c00kie17)\
**Post date:** [May 14, 2026, 12:00am UTC](https://community.fly.io/t/cant-restart-or-access-machine/27863/4 "2026-05-14T00:00:05Z")

</div>

It works now, thanks for the help and recommendation. I will setup a multiple machines to host the postgres instance

---

<div class="post-metadata">

**Author:** ![system](https://global.discourse-cdn.com/fly1/original/3X/6/e/6e8955d00cb218e186ebbe3a6ab06fefdce15290.svg) [@system](https://community.fly.io/u/system)\
**Post date:** [May 21, 2026, 12:00am UTC](https://community.fly.io/t/cant-restart-or-access-machine/27863/5 "2026-05-21T00:00:40Z")

</div>

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