From f439d18c5aa50142ae40f78f7d4da1cbe6df4538 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Fran=C3=A7ois=20Thir=C3=A9?= Date: Mon, 24 Feb 2025 17:12:50 +0100 Subject: [PATCH] Changelog --- CHANGES.rst | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/CHANGES.rst b/CHANGES.rst index aefce0899080..7d672cd0b5fb 100644 --- a/CHANGES.rst +++ b/CHANGES.rst @@ -49,6 +49,14 @@ Node ``/chain/{chain_id}/protocols/{protocol_hash}``) to retrieve protocol activation levels of the chain. (MR :gl:`!15447`) +- The node will detect stalled connections more quickly (on + Linux-based distributions). This behavior can be controlled via the + environment variable ``OCTEZ_P2P_TCP_USER_TIMEOUT``. Its default + value is ``15000``, meaning that it will now take ``15s`` to detect + a stalled connection (compared to up to ``15`` minutes by default on + Linux). Users can opt out by setting the value to ``0``. (MR + :gl:`!16907`) + Client ------ @@ -256,6 +264,15 @@ Data Availability Layer (DAL) DAL node ~~~~~~~~ +- **Feature** The node will detect stalled connections more quickly (on + Linux-based distributions). This behavior can be controlled via the + environment variable ``OCTEZ_P2P_TCP_USER_TIMEOUT``. Its default + value is ``15000``, meaning that it will now take ``15s`` to detect + a stalled connection (compared to up to ``15`` minutes by default on + Linux). Users can opt out by setting the value to ``0``. (MR + :gl:`!16907`) + + - **Feature** A new RPC ``/p2p/gossipsub/reconnection_delays`` which provides for each unreachable point, the time remaining until the next reconnection attempt. (MR :gl:`!16767`) -- GitLab