Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!mnetor!uunet!husc6!mit-eddie!ll-xn!ames!ucbcad!ucbvax!vms.CIS.PITTSBURGH.EDU!Postmaster From: Postmaster@vms.CIS.PITTSBURGH.EDU (PMDF Mail Server) Newsgroups: comp.os.vms Subject: Undeliverable mail Message-ID: <8710280102.AA14253@ucbvax.Berkeley.EDU> Date: Mon, 26-Oct-87 10:44:00 EST Article-I.D.: ucbvax.8710280102.AA14253 Posted: Mon Oct 26 10:44:00 1987 Date-Received: Fri, 30-Oct-87 06:48:36 EST Sender: daemon@ucbvax.BERKELEY.EDU Organization: The ARPA Internet Lines: 54 The message could not be delivered to: Addressee: 000426 Reason: %MAIL-E-NOSUCHUSR, no such user 10217_000426 at node CISVM2 ---------------------------------------- Received: from JNET-DAEMON by vms.cis.pittsburgh.edu; Mon, 26 Oct 87 01:11 EDT Received: From CMUCCVMA(MAILER) by PITTVMS with RSCS id 3721 for 000426@PITTVMS; Mon, 26-OCT-1987 00:29 EDT Received: by CMUCCVMA (Mailer X1.25) id 3719; Mon, 26 Oct 87 00:31:06 EST Date: Sat, 24 Oct 87 16:48 EDT From: "Clayton, Paul D." Subject: Re: Thoughts On Powering Up UNIBUS Boxes On Live Systems... Sender: INFO-VAX Discussion To: BETH DOE <000426@vms.cis.pittsburgh.edu> Reply-to: INFO-VAX@KL.SRI.COM Comments: To: INFO-VAX@kl.sri.com Information From TSO Financial - The Saga Continues... Chapter 30 - October 24, 1987 Chuck Rothauser of United Technologies Research Center had a trashed system disk after his local field service office powered up a UNIBUS on a LIVE system. The disk was located off an HSC and not a UDA interface so that cause can be ruled out. The data patterns that he listed in the message were similar in nature to the system disk I had trashed when my HSC CPU card went to lunch. I have not had a similar problem directly from the 78/8XXX processor. I have trashed disks when the FPA went to lunch but thats the extent of it. Chuck asked for comments. Other then the ones listed above, I add that I would NEVER let my local boys power up a UNIBUS on a live system. The problems are many fold. The ripple in power when its turned on, may be just enough to make another power supply go marginal in the output voltage its supplying to who knows what. The case of the device drivers being able to handle a power fail/initialization sequence may be questionable. Any third party interfaces may do strange things right after a power up and cause VMS to crash. The problem with LATs and having printers hang off the Ethernet would probably show up in a large way. The bottom line is that its NOT worth the problems to power up a UNIBUS on a live system in my mind. Hope this helps. :-) Paul D. Clayton - Manager Of Systems TSO Financial - Horsham, Pa. USA Address - CLAYTON%XRT@CIS.UPENN.EDU