From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from smtp40.i.mail.ru (smtp40.i.mail.ru [94.100.177.100]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by dev.tarantool.org (Postfix) with ESMTPS id 5607F42EF5C for ; Tue, 23 Jun 2020 11:39:28 +0300 (MSK) References: <46ed0cc08e6ecf7ee7c9250d9cdfc3d68aa5c764.1592589312.git.sergepetrenko@tarantool.org> <30f75001-3ecc-9d7f-9cba-88c824087f4e@tarantool.org> From: Serge Petrenko Message-ID: <6349f4ba-88c2-e6b4-1785-50e18dd5d010@tarantool.org> Date: Tue, 23 Jun 2020 11:39:26 +0300 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset="utf-8"; format="flowed" Content-Transfer-Encoding: 8bit Content-Language: en-GB Subject: Re: [Tarantool-patches] [PATCH 1/2] box: rework local_recovery to use async txn_commit List-Id: Tarantool development patches List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Vladislav Shpilevoy , sergos@tarantool.org, gorcunov@tarantool.org, lvasiliev@tarantool.org Cc: tarantool-patches@dev.tarantool.org 23.06.2020 00:43, Vladislav Shpilevoy пишет: > I am looking at the commit which added on_write_failure. > Why can't we reuse on_rollback trigger? As we discussed, > we can pass whatever we want using txn->signature. So > for manual rollback we can set it to -1, for write failure > to -2, for limbo rollback to -3, for timeout to -4, etc. > > I am just afraid that +16 bytes for transaction object > may affect performance of the async transactions. We can. I've just done it & dropped the additionnal commit (the one that hacked the trigger removal in recovery. -- Serge Petrenko