From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tommy Apel Hansen Subject: Re: raid1 read balance with write mostly Date: Sat, 02 Feb 2013 23:16:23 +0100 Message-ID: <1359843383.430.18.camel@workstation-home> References: <36edd03089332985af9a97eeaf719cd3@mail1.volny.cz> <510D8A8C.5090505@hardwarefreak.com> <20130203040758.5e2e7f3c@natsu> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-Reply-To: <20130203040758.5e2e7f3c@natsu> Sender: linux-raid-owner@vger.kernel.org To: Roman Mamedov Cc: stan@hardwarefreak.com, tomas.hodek@volny.cz, linux-raid@vger.kernel.org List-Id: linux-raid.ids On Sun, 2013-02-03 at 04:07 +0600, Roman Mamedov wrote: > On Sat, 02 Feb 2013 15:52:12 -0600 > Stan Hoeppner wrote: >=20 > > So under heavy read load, in this case we'd probably want the slow = 3TB drive > > to contribute to the workload. >=20 > No, it should behave as documented. >=20 > -W, --write-mostly > subsequent devices listed in a --build, --create, or --= add com=E2=80=90 > mand will be flagged as 'write-mostly'. This is valid = for RAID1 > only and means that the 'md' driver will avoid read= ing from > these devices if at all possible. ^^^^^^^^^^^^^^^^^^^ I would assume that if the kernel thinks the SSD is starved for io (tha= t being the %util column in iostat) it would consider sending io to the HDD. At least that would be what I would expect to happen, so if the SS= D reports wrong utilization numbers back or anything else involved in the io processing the the behavior is as expected. > =20 >=20 > if at all possible =3D if there is any other mirror still alive, do n= ot read from > this one. >=20 > BTW I also run an SSD+HDD array, and after a brief testing today I di= d notice > some reads from the HDD on reading from the array showing up in iosta= t. So I > can confirm the bug. >=20 /Tommy -- To unsubscribe from this list: send the line "unsubscribe linux-raid" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html