From owner-freebsd-hackers Fri Mar 10 00:02:46 1995 Return-Path: hackers-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id AAA08827 for hackers-outgoing; Fri, 10 Mar 1995 00:02:46 -0800 Received: from Root.COM (implode.Root.COM [198.145.90.1]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id AAA08804 for ; Fri, 10 Mar 1995 00:02:25 -0800 Received: from corbin.Root.COM (corbin.Root.COM [198.145.90.18]) by Root.COM (8.6.8/8.6.5) with ESMTP id AAA02174; Fri, 10 Mar 1995 00:01:54 -0800 Received: from localhost (localhost [127.0.0.1]) by corbin.Root.COM (8.6.9/8.6.5) with SMTP id AAA04134; Fri, 10 Mar 1995 00:01:54 -0800 Message-Id: <199503100801.AAA04134@corbin.Root.COM> X-Authentication-Warning: corbin.Root.COM: Host localhost didn't use HELO protocol To: hasty@netcom.com (Amancio Hasty Jr) cc: hackers@FreeBSD.org Subject: Re: SNAP or current? In-reply-to: Your message of "Thu, 09 Mar 95 23:22:10 PST." <199503100722.XAA06607@netcom14.netcom.com> From: David Greenman Reply-To: davidg@Root.COM Date: Fri, 10 Mar 1995 00:01:53 -0800 Sender: hackers-owner@FreeBSD.org Precedence: bulk >> Umm, I wouldn't count on the next SNAP being too stable. I have a dozen >>serious bugs (panics or system hangs) on the whiteboard, and I don't expect > >Care to elaborate ? > >Mostly because I am not seeing any panics over here (cross my fingers) and >perhaps they could be due to a driver or option configured in the kernel. Try copying a large file to a mounted msdos filesystem; it will destroy your root filesystem. Try using NQNFS; it will hang your machine. Try using an 8bit ethernet card, or a 16bit card in a slow machine w/heavy net traffic; your machine will panic. Try using bounce buffers; your machine will eventually panic in vm_bounce_alloc(). Try using your machine in a heavy load environment like freefall; it will eventually hang in a vnode lock deadlock. Try using quotas with multiple filesystems; your machine will panic. Try "ifconfig sl0 1.1.1.1; route add default 1.1.1.1; telnet 1.1.1.2"; your machine won't bother panicing, it'll just reboot because of a stack overflow. Try getting a soft error on your floppy (or try formatting one); it used to cause a panic, now it just prints out a nasty message telling us to fix the bug with biodone() being called more than once. ...and then there's the mmap() bugs that will hang your machine, etc, etc. The above is just a sample and doesn't include what I consider the most serious of the bugs - the wild pointer that randomly corrupts kernel memory. If I could reproduce that problem at will, I'd have it fixed by now. Unfortunately, it happens about 1 in 100 times I boot the machine, just after fsck finishes...or at random other times during normal system use. If you're lucky, it will just cause the machine to die. If you're unlucky like John, it will corrupt some FS buffers and destroy your superblock and root directory. Don't get to comfortable with -current...and do regular backups. :-( -DG