Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 15 Jun 2012 15:16:21 -0700
From:      Greg Byshenk <freebsd@byshenk.net>
To:        Kurt Jaeger <lists@opsec.eu>
Cc:        freebsd-stable@freebsd.org
Subject:   Re: Experience with Intel SATA and fbsd 8.3-amd64 ?
Message-ID:  <20120615221621.GS1768@portland1.byshenk.net>
In-Reply-To: <20120615172028.GU5335@home.opsec.eu>
References:  <20120608144621.GA74475@home.opsec.eu> <201206091527.q59FRTeg069656@lurza.secnetix.de> <20120615172028.GU5335@home.opsec.eu>

next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, Jun 15, 2012 at 07:20:28PM +0200, Kurt Jaeger wrote:
> > Kurt Jaeger <lists@opsec.eu> wrote:

> >  > I have a problem with some host: If I put heavy IO load on that
> >  > system, write errors happen, and then it crashes.
> > 
> > What kind of write errors, exactly?  What messages do you
> > get on the console?
> 
> g_vfs_done():ada0s1f[WRITE(offset=50699862016, length=16384)]error = 2
> 2
> g_vfs_done():ada0s1f[WRITE(offset=50699862016, length=16384)]error = 22
> g_vfs_done():ada0s1e[WRITE(offset=44693307392, length=16384)]error = 22
> g_vfs_done():ada0s1e[WRITE(offset=44693211136, length=2048)]error = 5
> 
> > It's also worth mentioning that such problems could also
> > be caused by bad RAM, or even by the power supply (though
> > the latter is unlikely in this case, I think).
> 
> Well, the device was probably a bit on the cheap side (ALLNET FW9000).

Could it be a device problem? I've seen that type of error
(including a crash in the end) when a device can't handle DMA.
Disabling DMA solved the problem for me.


-- 
greg byshenk  -  gbyshenk@byshenk.net  -  Leiden, NL - Portland, OR USA



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20120615221621.GS1768>