Date: Thu, 10 Jul 2014 14:58:22 +0300 From: Daniel Braniss <danny@cs.huji.ac.il> To: Steven Hartland <killing@multiplay.co.uk> Cc: FreeBSD-scsi <freebsd-scsi@freebsd.org> Subject: Re: Data corruption with the mfi(4) driver Message-ID: <A6573002-8FED-44A2-993B-9A65E4B11395@cs.huji.ac.il> In-Reply-To: <803159D4D0C44F879D21B6E0ECFE391F@multiplay.co.uk> References: <20140710092251.GA1206@sekishi.zefyris.com> <F9A3C574C3C64BC5A4DC61F5FDDD0342@multiplay.co.uk> <20140710105307.GD1206@sekishi.zefyris.com> <765694DA8A58468484623092946127D1@multiplay.co.uk> <20140710113315.GA3449@sekishi.zefyris.com> <803159D4D0C44F879D21B6E0ECFE391F@multiplay.co.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
On Jul 10, 2014, at 2:50 PM, Steven Hartland <killing@multiplay.co.uk> = wrote: > ----- Original Message ----- From: "Francois Tigeot" = <ftigeot@wolfpond.org> >=20 >=20 >> On Thu, Jul 10, 2014 at 12:24:21PM +0100, Steven Hartland wrote: >>> The key thing to determine here are precise factors of the = corruption your >>> seeing as it looks like there is some mixing up of issues >> My original mail was intended as a head's up notice; I don't have = enough >> information to fill in every detail and point to a particular root = cause so >> far. >>> Given this could you confirm:- >>> 1. Hardware >>> 1.1. LSI card / generation >> Dell Rx20 server generation >> I have personally seen the issue on a Dell R720xd with a LSI Dell = Perc H710 >> adapter. PCI Id: >> mfi0@pci0:3:0:0: class=3D0x010400 card=3D0x1f341028 chip=3D0x005b1000 = rev=3D0x05 >> hdr=3D0x00 but have had private reports from at least two different = machines >> having the same problem. >=20 > Perc H710 is 2208 based. >=20 >> All other parameters are variable. >=20 > They may be variable but need a point of reference, so detailing the = examples > is key. we have several machines with said PERCs, used as ZFS/NFS for postgres = to large dna sequences and have yet to receive complains of data corruption, this does not mean = the absence of them :-), but considering the amount of storage/servers some user should have = complained by now =85 so, is someone else in FreeBSD 9.2/3 seeing such corruptions? BTW, the only issue we had was when the server was connected at 10G we = did have a lot of data corruption which was solved by turning TSO off. danny >=20 > Regards > Steve > _______________________________________________ > freebsd-scsi@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-scsi > To unsubscribe, send any mail to = "freebsd-scsi-unsubscribe@freebsd.org"
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?A6573002-8FED-44A2-993B-9A65E4B11395>