Date: Mon, 21 Sep 1998 23:54:49 +0200 (CEST) From: Wilko Bulte <wilko@yedi.iaf.nl> To: ken@plutotech.com (Kenneth D. Merry) Cc: dnelson@emsphone.com, kuku@gilberto.physik.RWTH-Aachen.DE, freebsd-current@freefall.cdrom.com, freebsd-scsi@FreeBSD.ORG Subject: Re: ncr - new log messages appearing Message-ID: <199809212154.XAA02617@yedi.iaf.nl> In-Reply-To: <199809212020.OAA08872@panzer.plutotech.com> from "Kenneth D. Merry" at "Sep 21, 98 02:20:59 pm"
next in thread | previous in thread | raw e-mail | index | archive | help
As Kenneth D. Merry wrote... > > That just means your drive can't handle more than 15 tagged commands at > > once. CAM's max is 64 (is there any way to raise this?), so the first > > time your system hits the disk heavily, your drive will complain about > > the number of tags it can't handle, and the kernel message will pop up. > > You can adjust the maximum by adjusting the default quirk entry in > cam_xpt.c or by putting in a specific quirk entry for your device that sets > the limit higher. Most drives don't have any more than 64 tags. I think > there are some external RAID controllers that may have more. Quite true. For example the Digital HSZ40 controllers have a pool of 300 tags, that are shared by all LUNs that are configured on the array. There is one thing that must be kept in mind for those RAID boxes: you might encounter a QUEUE FULL even when that particular LUN had no I/O outstanding. Other I/O on different LUNs in that case has used up all tags from the pool. I've seen drivers choke on this before ;-) Wilko _ ______________________________________________________________________ | / o / / _ Bulte email: wilko@yedi.iaf.nl |/|/ / / /( (_) Arnhem, The Netherlands WWW : http://www.tcja.nl ______________________________________________ Powered by FreeBSD __________ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199809212154.XAA02617>