From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <alexander.turenko@tarantool.org>
Date: Sat, 20 Oct 2018 02:24:13 +0300
From: Alexander Turenko <alexander.turenko@tarantool.org>
Subject: Re: [PATCH v2 4/5] test: use wait_cond to check follow status
Message-ID: <20181019232413.3hmynv532fe2srbi@tkn_work_nb>
References: <20181019161721.49560-1-sergw@tarantool.org>
 <20181019161721.49560-5-sergw@tarantool.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <20181019161721.49560-5-sergw@tarantool.org>
To: Sergei Voronezhskii <sergw@tarantool.org>
Cc: tarantool-patches@freelists.org, Vladimir Davydov <vdavydov.dev@gmail.com>
List-ID: <tarantool-patches.dev.tarantool.org>

Hi!

See below.

WBR, Alexander Turenko.

On Fri, Oct 19, 2018 at 07:17:20PM +0300, Sergei Voronezhskii wrote:
> If `test_run:wait_cond()` found a not 'follow` status it returns true.
> Which immediately causes an error.
> 
> Fixes #3734
> Part of #2436, #3232
> ---
>  test/replication/misc.result   | 17 +++++++++++------
>  test/replication/misc.test.lua | 15 +++++++++------
>  2 files changed, 20 insertions(+), 12 deletions(-)
> 

> diff --git a/test/replication/misc.test.lua b/test/replication/misc.test.lua
> index 06ad974db..3866eb3ac 100644
> --- a/test/replication/misc.test.lua
> +++ b/test/replication/misc.test.lua
> @@ -53,15 +53,18 @@ fiber=require('fiber')
>  box.cfg{replication_timeout = 0.01, replication_connect_timeout=0.01}
>  _ = box.schema.space.create('test_timeout'):create_index('pk')
>  test_run:cmd("setopt delimiter ';'")
> +function wait_follow(replicaA, replicaB)
> +    return test_run:wait_cond(function()
> +        return replicaA.status ~= 'follow' or replicaB.status ~= 'follow'
> +    end, 0.01)
> +end ;
>  function test_timeout()
>      for i = 0, 99 do 
> +        local replicaA = box.info.replication[1].upstream or box.info.replication[2].upstream
> +        local replicaB = box.info.replication[3].upstream or box.info.replication[2].upstream
>          box.space.test_timeout:replace({1})
> -        fiber.sleep(0.005)
> -        local rinfo = box.info.replication
> -        if rinfo[1].upstream and rinfo[1].upstream.status ~= 'follow' or
> -           rinfo[2].upstream and rinfo[2].upstream.status ~= 'follow' or
> -           rinfo[3].upstream and rinfo[3].upstream.status ~= 'follow' then
> -            return error('Replication broken')
> +        if wait_follow(replicaA, replicaB) then
> +            return error(box.info.replication)

AFAIU, this test case checks that replicas do not leave from 'follow'
state even for a short time period. We should wait for 'follow' state
before the loop and perform some amount of attemps to catch an another
state. I don't sure, though. Georgy should draw the line.

I still think correction of test cases is a developer responsibility. If
you want to do it, please, discuss it with the author before. This will
save us some time we spend now on those extra review iterations.

>          end
>      end
>      return true
> -- 
> 2.18.0
>