From owner-freebsd-stable Fri Oct 13 17:48:33 2000 Delivered-To: freebsd-stable@freebsd.org Received: from mail.lewman.org (lowrider.lewman.org [63.109.230.166]) by hub.freebsd.org (Postfix) with ESMTP id BA69937B66E for ; Fri, 13 Oct 2000 17:48:27 -0700 (PDT) Received: by mail.lewman.org (Postfix, from userid 1001) id 273833D85; Fri, 13 Oct 2000 20:48:45 -0400 (EDT) Received: from localhost (localhost [127.0.0.1]) by mail.lewman.org (Postfix) with ESMTP id A5DD55BA0 for ; Fri, 13 Oct 2000 20:48:45 -0400 (EDT) Date: Fri, 13 Oct 2000 20:48:44 -0400 (EDT) From: Andy X-Sender: deimos@lowrider.lewman.org To: freebsd-stable@freebsd.org Subject: ata timeouts in 4.1-stable? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG I'm using fbsd 4.1-stable, last cvsup and make world was 2000-09-24. Everything looked fine until I added another ide drive, as secondary slave, to the primary master. I did check the archives first, and what I found was that people have reported this problem with ATA-66 controllers, not ATA-33. I now get log warnings such as: Oct 13 19:29:30 lowrider /kernel: ad2: WRITE command timeout - resetting Oct 13 19:29:30 lowrider /kernel: ata1: resetting devices .. done Oct 13 19:29:40 lowrider /kernel: ad2: WRITE command timeout - resetting Oct 13 19:29:50 lowrider /kernel: ata1: resetting devices .. done Oct 13 19:30:11 lowrider /kernel: ad2: WRITE command timeout - resetting Oct 13 19:30:11 lowrider /kernel: ata1: resetting devices .. done Oct 13 19:30:11 lowrider /kernel: ad2: WRITE command timeout - resetting Oct 13 19:30:11 lowrider /kernel: ata1: resetting devices .. done Oct 13 19:30:11 lowrider /kernel: ad2: WRITE command timeout - resetting Oct 13 19:30:11 lowrider /kernel: ata1-master: WARNING: WAIT_READY active=ATA_ACTIVE_ATA Oct 13 19:30:11 lowrider /kernel: ad2: trying fallback to PIO mode Oct 13 19:30:11 lowrider /kernel: ata1: resetting devices .. done My controller is: Oct 11 22:51:53 lowrider /kernel: atapci0: port 0xf000-0xf00f at device 7.1 on pci0 Oct 11 22:51:53 lowrider /kernel: ata0: at 0x1f0 irq 14 on atapci0 Oct 11 22:51:53 lowrider /kernel: ata1: at 0x170 irq 15 on atapci0 and my disks are: Oct 11 22:51:54 lowrider /kernel: ad0: 5006MB [10850/15/63] at ata0-master using UDMA33 Oct 11 22:51:54 lowrider /kernel: ad2: 4134MB [8960/15/63] at ata1-master using WDMA2 Oct 11 22:51:54 lowrider /kernel: acd0: CDROM at ata0-slave using PIO3 The errors above occur during the dump of an fs, the creation of a large tarball, or nightly run jobs that involve ad2. The actual disk in ad2 came out of a 3.5-stable system that ran without these errors. The disk appears to be fine, as I've put it back into the old system, and dumped the /usr fs to a tarball on it. These errors don't appear to be affecting the end result, as the tarballs, and dump files are still valid, but I'm wondering if these are cause for concern. Or if I should add this to a PR? -- | Andy | e-mail | web | | | andy@lewman.com | www.lewman.com | ... If forced to travel on an airplane, try and get in the cabin with the Captain, so you can keep an eye on him and nudge him if he falls asleep or point out any mountains looming up ahead ... -- Mike Harding, "The Armchair Anarchist's Almanac" To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message