Date: Thu, 27 May 2010 06:33:39 +0300 From: Alexander Motin <mav@FreeBSD.org> To: technews <technews@giallarhorn.org>, FreeBSD Stable <freebsd-stable@freebsd.org> Subject: Re: gmirror panic with SiI 3726 when array is degraded Message-ID: <4BFDE813.4090201@FreeBSD.org> In-Reply-To: <mailpost.1274922104.9085931.80471.mailing.freebsd.stable@FreeBSD.cs.nctu.edu.tw> References: <mailpost.1274922104.9085931.80471.mailing.freebsd.stable@FreeBSD.cs.nctu.edu.tw>
next in thread | previous in thread | raw e-mail | index | archive | help
technews wrote: > So I've started using gmirror recently to mirror 2 1.5 TB drives however > if I reboot the machine or 'gmirror stop -f data1' while the array is > been rebuilt I get the panic below. > Rebooting without geom_mirror loaded does not cause a panic > Rebooting with geom_mirror loaded and the drives disconnected does not > cause a panic. > The drives are in an "Rosewill RSV-S4-X 4 Bay SATA to eSATA (Port > Multiplier) JBOD / RAID 0, 1, 1+0, 5 Enclosure" connected over a SiI > 3726 (rev=1706) adapter > Rebooting with the drives mounted normally does not cause a panic > I'm using the latest and greatest firmware, the panics occurred before > the update It would be nice if you build kernel with debugger and got a stack backtrace to show where problem occurred. Also, if you like your port multiplier to work much faster and stable - I would strongly suggest you to update to 8-STABLE (or wait forthcoming 8.1-RELEASE, if you prefer releases) and use new siis(4) driver for your SiI3132 SATA controller. Port Multiplier support in ata(4) driver left in embryo level and won't be improved. -- Alexander Motin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4BFDE813.4090201>