From owner-freebsd-questions Thu Oct 9 06:39:55 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id GAA00435 for questions-outgoing; Thu, 9 Oct 1997 06:39:55 -0700 (PDT) (envelope-from owner-freebsd-questions) Received: from absinthe.i3inc.com (Absinthe.i3inc.com [209.31.147.194]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id GAA00421 for ; Thu, 9 Oct 1997 06:39:46 -0700 (PDT) (envelope-from chris@absinthe.i3inc.com) Received: from localhost (localhost [127.0.0.1]) by absinthe.i3inc.com (8.7.2/8.7.2) with SMTP id JAA13489; Thu, 9 Oct 1997 09:36:52 -0400 (EDT) Message-Id: <199710091336.JAA13489@absinthe.i3inc.com> X-Authentication-Warning: absinthe.i3inc.com: Host localhost [127.0.0.1] didn't use HELO protocol To: Tim Tsai Cc: questions@freebsd.org Subject: Broken Amanda/Archive DAT chio/seagate-changer X-Mailer: Mew version 1.03 on Emacs 19.34.1 Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Date: Thu, 09 Oct 1997 09:36:51 -0400 From: Chris Shenton Sender: owner-freebsd-questions@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In the archives, on May 6, 1997, you said about "amanda/Archive DAT changer problem solved": We've finally gotten Amanda working correctly with our Archive 4mm DAT changer (4586NP). First, 2.2/3.0 now has the chio SCSI changer code from NetBSD incorporated. Second, Amanda 2.3.0.4 from ftp.gps.caltech.edu includes a seagate-changer program that can control the changer directly using the chio mechanism. This has been a long standing problem for us and just thought that I point it out if somebody else is using a similar configuration. On that recommendation and a few others here recently, I got one from www.instantauction.com for $175. This changer holds 4 4mm DATs in an internal cassette with built-in changer mechanism. Seems happy to run with dump and manual amanda (I'm an amanda newbie but it looks like exactly what I need). Problem is I can't get the changer part to work. I'm running FreeBSD-2.2-Stable, cvsupped a week about a week ago. I compiled in support for ch0 in the kernel and it's recognized at boot: (ncr0:6:0): "ARCHIVE Python 28849-XXX 4.CM" type 1 removable SCSI 2 st0(ncr0:6:0): Sequential-Access st0(ncr0:6:0): 5.0 MB/s (200 ns, offset 8) density code 0x13, 512-byte blocks, write-enabled (ncr0:6:1): "ARCHIVE Python 28849-XXX 4.CM" type 8 removable SCSI 2 ch0(ncr0:6:1): Medium-Changer 4 slots, 1 drive, 1 picker Some of the chio commands don't work, most notably "status": operator@thanatos<17> chio status chio: /dev/ch0: CHIOGSTATUS: Input/output error After such a failure, the logs show the following, three times: Oct 9 09:19:38 thanatos /kernel: ch0(ncr0:6:1): extraneous data discarded. Oct 9 09:19:38 thanatos /kernel: ch0(ncr0:6:1): COMMAND FAILED (9 0) @f0642800. Others do work, including movement from drive to slot#, and slot# to drive: operator@thanatos<18> chio params /dev/ch0: 4 slots, 1 drive, 1 picker /dev/ch0: current picker: 0 operator@thanatos<19> chio move drive 0 slot 0 [does the right thing] operator@thanatos<20> chio move slot 3 drive 0 [does the right thing] I can't get *anything* to work with the Amanda "seagate-changer" command. I had to tweak it a little, adding "#ifdef HAVE_CHIO", changing the include to "", and changing the name of the function name "isnumber" to "myisnumber". Still, no joy -- IOCTL failures: operator@thanatos<26> seagate-changer -info ioctl failed: 0xffffffff Input/output error operator@thanatos<27> seagate-changer -reset ioctl failed: 0xffffffff Input/output error operator@thanatos<28> seagate-changer -eject ioctl failed: 0xffffffff Input/output error Syslog logs the same message, three times each, for each of the above command failures: Oct 9 09:25:19 thanatos /kernel: ch0(ncr0:6:1): extraneous data discarded. Oct 9 09:25:19 thanatos /kernel: ch0(ncr0:6:1): COMMAND FAILED (9 0) @f0642800. Can someone who's gotten this working tell me how you did it, or what I'm doing wrong that "chio status" and "seagate-changer -info" fail? Many thanks.