Date: Wed, 6 Aug 2003 10:53:28 -0400 From: Don Bowman <don@sandvine.com> To: "'Justin T. Gibbs'" <gibbs@scsiguy.com>, Don Bowman <don@sandvine.com>, "'freebsd-scsi@freebsd.org'" <freebsd-scsi@freebsd.org>, "'aic7xxx@freebsd.org'" <aic7xxx@freebsd.org> Subject: RE: Ongoing U320 AIC7902 Seagate ST318453LW issues, SCB timed out Message-ID: <FE045D4D9F7AED4CBFF1B3B813C8533702742163@mail.sandvine.com>
next in thread | raw e-mail | index | archive | help
From: Justin T. Gibbs [mailto:gibbs@scsiguy.com] > > On some systems I am sometimes getting this error pop out. > > It seems to track the system for some reason. This is > > not a bad cable, signals look good on oscope. > > > > Drive is: SEAGATE ST318453LW, FW 0005 > > You seem to have found another way to trip up the Seagate drives. > According to the controller, there are 13 commands that are > outstanding > on the drive that have not been completed within the timeout period. > The bus is idle and so is the controller. > > The SCB count in this dump is 64. Are you not throttling the number > of commands to 32 for these devices? If so, the SCB count should not > be higher than 32 for a single drive configuration. I have > seen behavior > similar to what you are experiencing with tag depths greater than 32. I'm continuing to test without the throttle. I'm @ a loss for why it tracks some systems and not others. There doesn't seem to be a reliable way to drop the number of tags since the system may not always come up. I don't think there's an option in the kernel to do so. --don
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?FE045D4D9F7AED4CBFF1B3B813C8533702742163>