mtfyun/ntfy
1
0 Fork 0
mirror of https://github.com/binwiederhier/ntfy.git synced 2026-11-10 14:44:04 +08:00
Code Issues Packages Projects Releases Wiki Activity

Release notes for v2.36.0

Browse Source
This commit is contained in:
binwiederhier committed 2026-10-31 15:00:21 +00:00
1 parent 2e8506f827
commit ea71722ba7
1 file changed
+31 -24
Download Patch File Download Diff File Expand all files Collapse all files
docs/releases.md
+31 -24
View File
@@ -6,12 +6,42 @@ and the [ntfy Android app](https://github.com/binwiederhier/ntfy-android/release
| Component | Version | Release date |
|------------------|---------|---------------|
| ntfy server | v2.35.0 | Oct 30, 2026 |
| ntfy server | v2.36.0 | Oct 31, 2026 |
| ntfy Android app | v1.25.2 | July 23, 2026 |
| ntfy iOS app | v1.7.0 | May 30, 2026 |
Please check out the release notes for [upcoming releases](#not-released-yet) below.
## ntfy server v2.36.0
Released October 31, 2026
This release makes the daily limits (messages, e-mails, phone calls) survive restarts and, with PostgreSQL, shared
between servers, so a visitor gets one daily allowance in total. Each visitor's daily period now starts at its own time
of day instead of midnight UTC. It also adds an experimental local admin socket, and checks the whole server config
again when `base-url` is set, so an invalid config that used to start now fails at startup.
**Features:**
* Daily limits (messages, e-mails, phone calls) are kept in the user database (`auth-file` for SQLite, or `database-url` for PostgreSQL) and survive restarts; with PostgreSQL, all servers share them, so a visitor gets one daily allowance in total, not one per server (see [message limits](config.md#message-limits))
* Experimental local admin socket (`experimental-listen-admin-unix`, off by default) that serves the admin API without credentials to whoever can open it (the ntfy user and group, and root); the `ntfy user/access/token/tier` commands do not use it yet. The shipped systemd unit now sets `RuntimeDirectory=ntfy`, so the server can create it in `/run/ntfy` (see [admin socket](config.md#admin-socket))
* Admin API (internal, undocumented): manage tokens of other users, tiers and password reset links
* Export `ntfy_db_up` (1 if the last health check of a PostgreSQL pool's database succeeded, per pool) and health-check the primary like the replicas, to alert on a server that is cut off from its database (see [PostgreSQL](config.md#postgresql))
**Behavior changes:**
* Daily limits now reset on a rolling 24-hour period per account or IP address instead of for everyone at midnight UTC: each visitor's period starts at its own time of day, which spreads the resets over the day. The account page and the `/v1/account` API (`stats.reset_at`) show when the visitor's counts reset
* Visitors without a tier have a daily e-mail limit again (24h / `visitor-email-limit-replenish`), now that it is shared between servers and survives restarts
* The daily reset job is gone; the `stats_*` columns of the user table are zeroed by the upgrade and no longer read or written, so everyone's daily counts start over at the upgrade
* Servers that share a PostgreSQL database must keep their clocks in sync (NTP): the daily periods and the usage sync use each server's own clock
**Bug fixes + maintenance:**
* Admin API: return 4xx errors instead of a 500 for invalid input, and remove a deleted user's web push subscriptions
* Web app: Deleted notifications no longer come back when a poll replays messages published in the same second as the last one ([#2098](https://github.com/binwiederhier/ntfy/issues/2098), thanks to [@mattford63](https://github.com/mattford63) and [@arminus](https://github.com/arminus) for reporting)
* Web app: A message replayed after a reconnect keeps its read state and no longer notifies again, and an older version of a sequence no longer replaces a newer one ([#2090](https://github.com/binwiederhier/ntfy/issues/2090), thanks to [@tobykaspar79](https://github.com/tobykaspar79) for reporting)
* Check the whole server config again when `base-url` or a `message-size-limit` above 4K is set; most checks after them were skipped, so an invalid config (e.g. `upstream-base-url` equal to `base-url`) now fails to start ([#1969](https://github.com/binwiederhier/ntfy/pull/1969), thanks to [@mikejgray](https://github.com/mikejgray) for the contribution)
* CLI: `ntfy pub --wait-pid` now waits for processes owned by other users instead of failing with "not running" ([#1710](https://github.com/binwiederhier/ntfy/pull/1710), thanks to [@bekoeppel](https://github.com/bekoeppel) for the contribution)
## ntfy server v2.35.0
Released October 30, 2026
@@ -2288,29 +2318,6 @@ and the [ntfy Android app](https://github.com/binwiederhier/ntfy-android/release
## Not released yet
### ntfy server v2.36.0 (UNRELEASED)
**Features:**
* Daily limits (messages, e-mails, phone calls) are kept in the user database (`auth-file` for SQLite, or `database-url` for PostgreSQL) and survive restarts; with PostgreSQL, all servers share them, so a visitor gets one daily allowance in total, not one per server (see [message limits](config.md#message-limits))
* Local admin socket (`listen-admin-unix`, default `/run/ntfy/admin.sock`) that serves the admin API without credentials to whoever can open it (the ntfy user and group, and root); the shipped systemd unit now sets `RuntimeDirectory=ntfy` so the server can create it. Custom unit files without it log a warning at startup and keep working as before (see [admin socket](config.md#admin-socket))
* Admin API (internal, undocumented): manage tokens of other users, tiers and password reset links
* Export `ntfy_db_up` (1 if the last health check of a PostgreSQL pool's database succeeded, per pool) and health-check the primary like the replicas, to alert on a server that is cut off from its database (see [PostgreSQL](config.md#postgresql))
**Behavior changes:**
* Daily limits now reset on a rolling 24-hour period per account or IP address instead of for everyone at midnight UTC: each visitor's period starts at its own time of day, which spreads the resets over the day. The account page and the `/v1/account` API (`stats.reset_at`) show when the visitor's counts reset
* Visitors without a tier have a daily e-mail limit again (24h / `visitor-email-limit-replenish`), now that it is shared between servers and survives restarts
* The daily reset job is gone; the `stats_*` columns of the user table are zeroed by the upgrade and no longer read or written
**Bug fixes + maintenance:**
* Admin API: return 4xx errors instead of a 500 for invalid input, and remove a deleted user's web push subscriptions
* Web app: Deleted notifications no longer come back when a poll replays messages published in the same second as the last one ([#2098](https://github.com/binwiederhier/ntfy/issues/2098), thanks to [@mattford63](https://github.com/mattford63) and [@arminus](https://github.com/arminus) for reporting)
* Web app: A message replayed after a reconnect keeps its read state and no longer notifies again, and an older version of a sequence no longer replaces a newer one ([#2090](https://github.com/binwiederhier/ntfy/issues/2090), thanks to [@tobykaspar79](https://github.com/tobykaspar79) for reporting)
* Check the whole server config again when `base-url` or a `message-size-limit` above 4K is set; most checks after them were skipped, so an invalid config (e.g. `upstream-base-url` equal to `base-url`) now fails to start ([#1969](https://github.com/binwiederhier/ntfy/pull/1969), thanks to [@mikejgray](https://github.com/mikejgray) for the contribution)
* CLI: `ntfy pub --wait-pid` now waits for processes owned by other users instead of failing with "not running" ([#1710](https://github.com/binwiederhier/ntfy/pull/1710), thanks to [@bekoeppel](https://github.com/bekoeppel) for the contribution)
### ntfy iOS app v1.8.0 (UNRELEASED)
**Features:**