Path: utzoo!utgpu!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!mailrus!cornell!uw-beaver!rice!sun-spots-request From: SYSRUTH@UTORPHYS.BITNET (Ruth Milner) Newsgroups: comp.sys.sun Subject: Re: problems with Xylogics 451/CDC Sabre 850 Message-ID: <8812201900.AA05132@gpu.utcs.toronto.edu> Date: 30 Dec 88 19:22:06 GMT Sender: usenet@rice.edu Organization: Sun-Spots Lines: 35 Approved: Sun-Spots@rice.edu Original-Date: Tue, 20 Dec 88 14:00 EST X-Sun-Spots-Digest: Volume 7, Issue 82, message 7 of 18 X-Issue-Reference: v7n53 In v7n53, (Bradford Garton) writes: >We've been having mucho problems with a Xylogics 451/CDC Sabre 850 >combination. Here's the story (for those interested): I was told by our vendor here (Dilog) that the 451 is not capable of handling the transfer rate of the bigger Sabre disks, 850 and 1230. They are simply too fast for it. This would certainly explain some of the problems you are seeing. Even when there are no user processes running, it is possible for system processes such as sendmail and uucp, and incoming Usenet news, to generate a fair amount of disk traffic. If this disk is NFS-served to other systems, someone elsewhere could have been requesting a big transfer. Intermittent problems could show up as a result of intermittent traffic on the disk. We have ordered a Xylogics 753 to handle our Sabre 1230. A Ciprico would work as well, although make sure you have the latest rev. as the earlier ones had some problems with "sweep mode" on the disks. I can't tell you yet how well it works as they haven't arrived :-) . You should query AVIV about this. I have dealt with them before, and found them to be very good on their technical backup, but it is possible that either they didn't understand the Sabre was going on the 451, or that your particular rep. didn't know this (or, of course, that the information I was given wasn't correct; but the 850 and 1230 have a 16ms access time as opposed to 18ms on the smaller models, so it does make sense). Hope this helps. Ruth Milner Systems Manager University of Toronto Physics sysruth@helios.physics.utoronto.ca (Internet) sysruth@utorphys (BITNET)