# PostgreSQL 15 update

**URL:** https://meta.discourse.org/t/postgresql-15-update/349515
**Category:** Announcements
**Created:** [January 29, 2025, 7:20am UTC](https://meta.discourse.org/t/postgresql-15-update/349515 "2025-01-29T07:20:00Z")
**Posts on this page:** 1
**Showing post:** 69

<div class="post-metadata">

### Author: ![mwaniki](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/mwaniki/32/313097_2.png) [@mwaniki](https://meta.discourse.org/u/mwaniki)
#### Post date: [February 3, 2025, 7:32am UTC](https://meta.discourse.org/t/postgresql-15-update/349515/69 "2025-02-03T07:32:13Z")

</div>

> [@schneeland](#):
>
> So I guess the database did not shutdown correctly and was forcibly shutdown after timeout?

It would seem so. Could you try to follow the steps in [this previous post](https://meta.discourse.org/t/postgresql-15-update/349515/67) and see if there are any other client connections to the database? Ideally, you should stop any other applications that connect to the database before stopping the `app` container.

Alternatively, you can try to terminate all established database sessions and quickly stop the `postgres` service (ideally before the client apps reconnect) then attempt another rebuild after confirming a clean database shutdown from the logs. However, I _strongly_ recommend that you first identify the impact on your client applications listed in `pg_stat_activity` before terminating their connections.

Here’s a sample command you can run to terminate client connections and stop `postgres` from inside the `app` container **after** stopping `nginx` and `unicorn` first.

```plaintext
sudo -u postgres psql -c "SELECT pg_terminate_backend(pid) FROM pg_stat_activity WHERE pid <> pg_backend_pid();" && sv stop postgres

```

---

_[View the full topic](https://meta.discourse.org/t/postgresql-15-update/349515)._