From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from localhost (localhost [127.0.0.1]) by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTP id 058C6302E2 for ; Thu, 6 Jun 2019 23:14:45 -0400 (EDT) Received: from turing.freelists.org ([127.0.0.1]) by localhost (turing.freelists.org [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id P8cEXbgNLgec for ; Thu, 6 Jun 2019 23:14:44 -0400 (EDT) Received: from smtpng2.m.smailru.net (smtpng2.m.smailru.net [94.100.179.3]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTPS id B6BEE2FE9E for ; Thu, 6 Jun 2019 23:14:44 -0400 (EDT) Date: Fri, 7 Jun 2019 06:14:23 +0300 From: Alexander Turenko Subject: [tarantool-patches] Re: [PATCH v1] test: use unix sockets iproto Message-ID: <20190607031420.aleakgwmhaklkfpc@tkn_work_nb> References: MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: Sender: tarantool-patches-bounce@freelists.org Errors-to: tarantool-patches-bounce@freelists.org Reply-To: tarantool-patches@freelists.org List-Help: List-Unsubscribe: List-software: Ecartis version 1.0.0 List-Id: tarantool-patches List-Subscribe: List-Owner: List-post: List-Archive: To: "Alexander V. Tikhonov" Cc: tarantool-patches@freelists.org I'm ok with the change itself, so formally LGTM. Please, proceed further with Kirill. WBR, Alexander Turenko. > test: use unix sockets iproto Typo: iproto -> for iproto. Maybe better 'test: use unix sockets for iproto connections'. First of all I would state explicitly 'why': "eliminated most of 'address already in use' errors seen from time to time in our testing". Then you can describe the reason of those fails, provide information about previous tries to solve it, give needed background and so on. > Enabled use_unix_sockets and use_unix_sockets_iproto options use_unix_sockets is already enabled, the phrase is misleading. > to use unix sockets iproto instead of TcpPortDispatcher which > was previously introduced to eliminate the race condition when > two workers trying to use the same port: the idea was that each > worker used its own ports range. Really these ports could race > with client ports (from, say, net.box or replication), which > typically didn't use bind() and so bound to a random available > port (despite any dispatched ranges) and could produce > 'Address already in use' error. I would separate a description of the new way to handle the problem ('Enabled use_unix_sockets_iproto option to let test-run appoint unix sockets for LISTEN environment variable values') and a description of previous approaches ('TCP port ranges and so on') if you really want to give enough context and describe the latter one entirely. It is not obligatory I think. I mean, now it is not easy to understand that you describe disadvantages on a *previous* approach and almost all this paragraph is not about a code that is enabled in this commit. I think it can mislead someone if (s)he has less knowledge around test-run then you. You are formally right however, so ignore my wording nitpicking if you don't agree and just proceed with Kirill. > > Closes: #4008 > --- > > Github: https://github.com/tarantool/tarantool/tree/avtikhon/gh-4008-test-in-parallel-iproto > Issue: https://github.com/tarantool/tarantool/issues/4008