From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <korablev@tarantool.org>
Received: from smtp55.i.mail.ru (smtp55.i.mail.ru [217.69.128.35])
 (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 02E484765E0
 for <tarantool-patches@dev.tarantool.org>;
 Thu, 24 Dec 2020 19:59:38 +0300 (MSK)
Date: Thu, 24 Dec 2020 16:59:37 +0000
From: Nikita Pettik <korablev@tarantool.org>
Message-ID: <20201224165937.GD14559@tarantool.org>
References: <cover.1608159414.git.lvasiliev@tarantool.org>
 <54ef9ccd2a7bee1f5f53a811c7edea1ba034c4ca.1608159414.git.lvasiliev@tarantool.org>
 <8e66938a-e950-1b80-e664-13af405dfaeb@tarantool.org>
 <200e5cd6-6c8c-d784-42ae-0a941a7b9e1b@tarantool.org>
 <9af9fff1-81d7-f7ce-ca7e-171f551dfc72@tarantool.org>
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <9af9fff1-81d7-f7ce-ca7e-171f551dfc72@tarantool.org>
Subject: Re: [Tarantool-patches] [PATCH v3 1/2] sql: add missing diag_set on
 failure when working inside os_unix.c
List-Id: Tarantool development patches <tarantool-patches.dev.tarantool.org>
List-Unsubscribe: <https://lists.tarantool.org/mailman/options/tarantool-patches>, 
 <mailto:tarantool-patches-request@dev.tarantool.org?subject=unsubscribe>
List-Archive: <https://lists.tarantool.org/pipermail/tarantool-patches/>
List-Post: <mailto:tarantool-patches@dev.tarantool.org>
List-Help: <mailto:tarantool-patches-request@dev.tarantool.org?subject=help>
List-Subscribe: <https://lists.tarantool.org/mailman/listinfo/tarantool-patches>, 
 <mailto:tarantool-patches-request@dev.tarantool.org?subject=subscribe>
To: Vladislav Shpilevoy <v.shpilevoy@tarantool.org>
Cc: tarantool-patches@dev.tarantool.org

On 24 Dec 17:00, Vladislav Shpilevoy wrote:
> Hi! Thanks for the fixes!
> 
> See 1 comment below. After that the patchset will LGTM, and
> you can send it to Nikita on a second review.

Right away I can say that it would be super nice to see corresponding test
(using error injections I gueess). Otherwise we can't say for sure that
all this stubs are working...
 
> > diff --git a/src/box/sql/os_unix.c b/src/box/sql/os_unix.c
> > index d64f1bd..07461f5 100644
> > --- a/src/box/sql/os_unix.c
> > +++ b/src/box/sql/os_unix.c
> > @@ -873,6 +908,12 @@ unixWrite(sql_file * id, const void *pBuf, int amt, sql_int64 offset)
> >              /* lastErrno set by seekAndWrite */
> >              return -1;
> >          } else {
> > +            /* Wrote to file less than planned. */
> > +            const char *err = tt_sprintf("wrote less (%i bytes) "
> > +                             "than planned (%i bytes) "
> > +                             "to file: %s",
> > +                             wrote, amt, pFile->zPath);
> > +            diag_set(ClientError, ER_SQL_EXECUTE, err);
> 
> Planned wasn't amt, because amt is modified in the loop above. Also
> 'wrote' will be -1 here, because this branch is taked on ENOSPC and
> write() returned -1.
> 
> Perpahs it is enough to say that didn't have enough space to write
> 'amt' bytes (this is how much was left to write).
> 
> >              storeLastErrno(pFile, 0);    /* not a system error */
> >              return -1;
> >          }