Date: Wed, 15 Dec 2010 16:27:55 +0100 From: Ivan Voras <ivoras@freebsd.org> To: freebsd-current@freebsd.org Cc: freebsd-geom@freebsd.org Subject: Re: Low speed disk subsystem after the upgrade 8.1-STABLE-201011 --> 9-CURRENT Message-ID: <ieampi$8ro$1@dough.gmane.org> In-Reply-To: <4D08DBC5.30003@freebsd.org> References: <4D08B21B.5090000@ukr.net> <4D08BC09.3000801@freebsd.org> <4D08BE7D.2030700@ukr.net> <4D08CC18.7050908@ukr.net> <AANLkTimwDePpdFZiaXS_5QM1W-wf%2BHh6Spcf907pud6%2B@mail.gmail.com> <4D08DBC5.30003@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 15.12.2010 16:16, Andriy Gapon wrote: > on 15/12/2010 16:25 Olivier Smedts said the following: >> 2010/12/15 Vladislav V. Prodan <universite@ukr.net>: >>> 15.12.2010 15:11, Vladislav V. Prodan wrote: >>> >>>> Now try to attach a screenshot. >>>> >>>> >>> >>> http://img576.imageshack.us/i/dsc00563u.jpg/ >> >> ad6: TIMEOUT - READ_DMA retrying (1 retry left) LBA=8638372 >> >> Maybe installkernel/installworld copied files on bad clusters, they >> were previously bad but you notice only now that there are system >> files on them... > > This indeed looks like a problem caused by disk. > There shouldn't have been the panic though ("bio_completed can't be greater than > bio_length"). > It's a possible geom or ad driver problem. > What is your geom topology on this system? > > Changing cc: from fs@ to geom@. Agreed, looks like hardware or driver issue, but AFAIK 8 and HEAD have pretty much the same ATA drivers, MFC-ed often. Maybe it's something trivial - nudged cables or a power supply gone accidentally bad. Just for a test, can you try running with AHCI here? (ahci_load=YES in loader.conf; note that it will rename devices; don't know if ZFS root can handle it).
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?ieampi$8ro$1>