From owner-freebsd-bugs Tue Nov 7 06:57:29 1995 Return-Path: owner-bugs Received: (from root@localhost) by freefall.freebsd.org (8.6.12/8.6.6) id GAA21004 for bugs-outgoing; Tue, 7 Nov 1995 06:57:29 -0800 Received: from ast.com (irvine.ast.com [165.164.128.2]) by freefall.freebsd.org (8.6.12/8.6.6) with SMTP id GAA20999 ; Tue, 7 Nov 1995 06:57:26 -0800 Received: from fw.ast.com by ast.com with SMTP id AA05744 (5.67b/IDA-1.5); Tue, 7 Nov 1995 06:58:41 -0800 Received: from nemesis by fw.ast.com with uucp (Smail3.1.29.1 #4) id m0tCpID-00008NC; Tue, 7 Nov 95 08:46 CST Received: by nemesis.lonestar.org (Smail3.1.27.1 #19) id m0tCpGU-000JD4C; Tue, 7 Nov 95 08:44 WET Message-Id: Date: Tue, 7 Nov 95 08:44 WET To: bugs@freebsd.org From: uhclem%nemesis@fw.ast.com (Frank Durda IV) Sent: Tue Nov 7 1995, 08:44:42 CST Cc: jkh@freebsd.org Subject: Bugs in/near installation 1104 SNAP Sender: owner-bugs@freebsd.org Precedence: bulk I monitored an installation on SNAP 1104 by watching the ALT-F2 display carefully, and there is a bunch of stuff going on that doesn't look right. 1. The hard disk in use is a Caviar 31200 IDE Manufacturer Boilerplate Label is 2484 cylinders, 16 heads, 63 cylinders, 1281.9MB CMOS setup (old BIOS, no EIDE) is 1023 cylinders, 16 heads, 63 cyls 180M DOS/Lose '95 partition with remainder of the disk used by FreeBSD. The system already has a DOS partition, and looks like this in the disk partition utility: 0 63 62 unused 0 63 368865 368927 wd0s1 fat 6 368928 2134944 2503871 wd0s2 freebsd 165 The drive was "sliced" during installtion as: wd0s1 180M DOS /dos wd02a 97MB UFS / y (newFS) wd02b 39MB SWAP wd02e 905MB UFS /usr y (newFS) That all seems reasonable. 1a. However, when the install gets going, the ALT-F2 display reports: ... DEBUG: Makedev: Called with wd0 on path /mnt/dev DEBUG: Makedev: Unit 0, Slice 1, Part 2 **** DEBUG: Makedev: Called with - on path /mnt/dev **** DEBUG: Makedev: - is an unknown device type! DEBUG: Makedev: Called with wd0s1 on path /mnt/dev ... The mystery "-" looks wrong. 1b. The install proceeds a bit further (or was it earlier) and then reports: **** Warning: calculated sectors per cylinder (4000) disagrees with **** disk label (1008) **** /dev/rwd0a 200000 sectors in 50 cylinders of 1 tracks, 4000 sectors **** 97,7MB in 4 cyl groups (16 c/g, 31.25MB/g, 3776 i/g) Isn't this a totally horrible and inefficient way to manage a disk? Plus I only get 2 or 3 alternate superblocks for a ~100MB partition. Only four cylinder groups? FYI, 63*16==1008, so the disk label is correct. 4000 is a mystery value and I have no idea where it came from. 2. This may not be a problem, but it also puts out: ... cpio: /mnt/dev/wd0s2a not created: newer or same age version exists **** 0 blocks ... Did something not extract correctly? 3. During kernel -c, if you use "visual" and specify an interrupt value of "2", it is not re-mapped to "9" as it is in the non-visual configuration utility. This makes strange things happen later on, particularly if you are setting the interrupt for ed0. 4. If you elect to configure X during install, someone needs to do a "stty erase ^H" before running the X configure script. 5. /bin/797 is still broken. Probeonly can't find the "X" you supposedly just installed and fails. 6. Probably not a bug, but without enabling any extra debug messages, the F2 screen randomly displays dozens of "DEBUG: FTP Close Called" messages throughout the extraction. 7. If remote site is unavailable when you try to download lynx during install, you get a "not on remote site" error, which is OK. After waiting for remote site to come back up, I asked the install to download lynx again. It failed and network board lights show it didn't even hit the network before saying it failed. But, if you ask for it once again, it hits the network, finds the files and downloads OK. 8. After rebooting, /etc/hosts has some trailing trash on the local system entry: (in my case) 165.164.6.19 skaro.lonestar.org skaro 165 *** The "165" probably should not be there. 9. Again, probably not a bug, but installing the compats leaves you with a "telnet" that won't run. telnet 165.164.6.15 ld.so failed: Undefined symbol "_dst_realm_sz" in telnet:telnet The "bad" telnet is about 10K larger than the "good" one. I don't recall reading in the installation that loading these modules would replace working modules with non-working ones, or it wasn't clear. Looks good otherwise! Frank Durda IV |"The Knights who say "LETNi" or uhclem%nemesis@fw.ast.com (Fastest Route)| demand... A SEGMENT REGISTER!!!" ...letni!rwsys!nemesis!uhclem |"A what?" ...decvax!fw.ast.com!nemesis!uhclem |"LETNi! LETNi! LETNi!" - 1983 ======= End Forwarded Message =================================================