Date: Sun, 12 Mar 2017 20:48:50 -0400 From: Mark Saad <nonesuch@longcount.org> To: Michael Fuckner <michael@fuckner.net> Cc: hackers@freebsd.org Subject: Re: Mini sas forward/ reverse cables ? Message-ID: <A1D22963-BAD8-4E47-B7ED-A0450DF9A7F2@longcount.org> In-Reply-To: <ca819d48-7aea-1485-4ca6-ea11847bb04b@fuckner.net> References: <4AF2C082-2494-42DE-B2C2-9C750BF39F76@longcount.org> <ca819d48-7aea-1485-4ca6-ea11847bb04b@fuckner.net>
next in thread | previous in thread | raw e-mail | index | archive | help
> On Mar 12, 2017, at 5:02 PM, Michael Fuckner <michael@fuckner.net> wrote: >=20 >> On 3/12/2017 9:54 PM, Mark Saad wrote: >> All >> I read this post on zfs corruption and there is an interesting comment a= bout cable types . >>=20 >> http://changelog.complete.org/archives/9769-silent-data-corruption-is-rea= l >> The part of the story I found interesting was how the disks didn't show u= p in the controller bios . > > I have that issue with a similar card on this server . So has anyone > e= ver seen this cabling issue ? Is it worth changing? How do I know > > which cable I have ? >=20 >=20 > it's pretty straight forward- if your controller is multilane and backplan= e single- you need a straight cable. This is the default! >=20 > If your controller is single and your backplane is multilane, your cable n= eeds to be cross over. If it is working, there is no need to change. >=20 What sort of markings would I look for ? Are the connections different ? Als= o the post indicated it was something they overlooked on their initial build= out that gave them a working ( albeit poorly ) setup . > Regards, > Michael! >=20 >=20 >=20 --- Mark Saad | nonesuch@longcount.org=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?A1D22963-BAD8-4E47-B7ED-A0450DF9A7F2>