Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!mnetor!uunet!husc6!mit-eddie!ll-xn!ames!ucbcad!ucbvax!VENUS.YCC.YALE.EDU!leichter From: leichter@VENUS.YCC.YALE.EDU ("Jerry Leichter") Newsgroups: comp.os.vms Subject: re: DECNET dies and users are disconnected Message-ID: <8709291556.AA19119@ucbvax.Berkeley.EDU> Date: Mon, 28-Sep-87 16:34:00 EDT Article-I.D.: ucbvax.8709291556.AA19119 Posted: Mon Sep 28 16:34:00 1987 Date-Received: Thu, 1-Oct-87 03:53:53 EDT Sender: daemon@ucbvax.BERKELEY.EDU Reply-To: "Jerry Leichter" Distribution: world Organization: The ARPA Internet Lines: 49 DECnet event 4.18, adjacency down Circuit UNA-0, dropped by adjacent node DECnet event 4.7, circuit down, circuit fault Circuit DMC-0, Line synchronization lost DECnet event 4.10 circuit up Circuit UNA-0 DECnet event 4.10 circuit up Circuit DMC-0 DECnet event 4.15 adjancency up Circuit UNA-0 I keep getting the above messages on the console once or twice a day. At the same time, all users are disconnected from that node. Some of the users are on a DZ11 and some are over ETHERNET. The node doing this is a part of a two node homogenous cluster. This problem does not occur on the other node at all. However, the node causing the problem is an area router for DECnet. I called the Remote Diagnostics Center about this. They said "software problem" and transferred me over to software support. Software support told me to increase NPAGEDYN, LRPCOUNT, and LRPCOUNTV. I did so, ran AUTOGEN, and rebooted. That didn't solve the problem. I have used SHOW MEMORY right after one of these mass disconnections has happened and I am convinced that it is not a memory problem as DEC said it was. Sounds like a hardware problem to me. Are the DZ11, DMC, and DEUNA on the same Unibus? Is there enough power available to that Unibus? Did the prob- lems by any chance start after you added some new device to the configuration? The "common elements" among the three failures you are seeing are either some fairly broadly-based software problem - I suppose it could be memory, though I would have expected to see rather more complaining from VMS if it was really running out - and any shared hardware. Take a look at your error logs, BTW; they should show something (probably errors reported against the UBA) if it is, indeed, the hardware that is at fault. Another clue to look for: Do users report "DAP CRC checksum errors" when transfering files through this machine? Unibus problems often present them- selves in this way.... -- Jerry ------