Date: Thu, 27 May 1999 12:33:49 -0500 From: Karl Denninger <karl@Denninger.Net> To: mjacob@feral.com, Dirk-Willem van Gulik <dirkx@webweaving.org>, Darryl Okahata <darrylo@sr.hp.com> Cc: freebsd-scsi@freebsd.org Subject: Re: OnStream SCSI drives Message-ID: <19990527123349.B50895@Denninger.Net> In-Reply-To: <Pine.BSF.4.05.9905271000550.64523-100000@semuta.feral.com>; from Matthew Jacob on Thu, May 27, 1999 at 10:11:18AM -0700 References: <Pine.BSF.4.05.9905251235340.360-100000@brunte.ispra.webweaving.org> <Pine.BSF.4.05.9905271000550.64523-100000@semuta.feral.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Thanks - I have decided to wait, as I found someone with two DLT 15/30s that they weren't using and got them REALLY cheap. I still want the Onstream support, but its no longer a fire. Now I know why the hacks I did (including the 32k fixed blocksize) didn't work. -- -- Karl Denninger (karl@denninger.net) Web: fathers.denninger.net I ain't even *authorized* to speak for anyone other than myself, so give up now on trying to associate my words with any particular organization. On Thu, May 27, 1999 at 10:11:18AM -0700, Matthew Jacob wrote: > > I got the documentation for this device and got a drive unit, and took a > quick look at this today (I have to be gone until next tuesday- my flight > leaves in a couple of hours). > > The support for this is going to be quite complex. Not only is it the > various modifications to commands (must have 'immediate' bit set, etc..), > but it means that: > > + All records are 32KBytes user data *PLUS* a 512 byte AUX field > that describes this tape frame. The AUX field has to not only > describe what's in this record but also record the number of > filemark frames preceeding this, etc... > > + Write errors due to defective media have to be handled in > software- i.e., the data relocated to the next frame, with > possibly updating the failed frame so that later read of that > frame knows that it's bogus (and where to go for the 'good' > frame). > > It will be possible to support variable mode or different blocksizes with > this tape format- if we want to. Basically we have to do in software that > which the Tandberg QIC tape devices do (and do in somwhat pixielike manner > at that..). In any case, this is not a no-brainer trivial task. I'll start > tackling it next week, but it'll probably take a couple of weeks at least. > > I'll ask Onstream if they mind the information going out so if others have > more aggressive schedules they can try to implement this sooner if they > lke, but you'd better coordinate with me on this. > > -matt > > > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19990527123349.B50895>