Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.3 4.3bsd-beta 6/6/85; site ucbvax.ARPA
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!houxm!ihnp4!ucbvax!daemon
From: info-vax@ucbvax.ARPA
Newsgroups: fa.info-vax
Subject: Dual path DSA disks.
Message-ID: <9929@ucbvax.ARPA>
Date: Fri, 16-Aug-85 01:29:41 EDT
Article-I.D.: ucbvax.9929
Posted: Fri Aug 16 01:29:41 1985
Date-Received: Mon, 19-Aug-85 07:42:53 EDT
Sender: daemon@ucbvax.ARPA
Organization: University of California at Berkeley
Lines: 28

From: Thomas J. Linscomb 

> From: ken@CIT-Hamlet.ARPA
> Subject:  Duel path DSA devices
> To: info-vax@CIT-Hamlet.ARPA
> 
> 	I've read that you can have two HSC50s serving the same DSA disk
> (for redundancy, only one HSC does the work at any one time). Has anyone
> tried to connect the second port of an HSC50 served disk to the UDA-50
> on one of the cluster vaxes? Presumably you'd need to set the allocation
> class the same and prey (or otherwise coerce) that the HSC50 was used as
> the primary path. I'm interested in this principally for robustness
> since we have been having trouble with our HSC50 (Circuit breaker in the
> back trips occasionally).
> -Ken Adelman
>  Caltech

  By design, dual porting a UDA disk does not work.  But I have sat with
someone at DECUS who had a system stay up after the HSC-50 serving the
system disk went away and it continued to run on the UDA-50.  Some people
are lucky.
  Remember all the smarts they put out in the UDA-50? What happens to all
those buffers and caches when you dual port the UDA?  They screw you around!
  As a note, I have heard that this is on Digital's wish list (reference
March 1983 PAGESWAPPER, p 18).  Also this does not prevent you from using
the UDA as a hot backup for the HSC.
--Thomas aka linscomb@ut-ngp
  The University of Texas Computation Center