On 2012/04/16 18:40 (GMT+0800) joeyli composed:
On 04/14/2012 02:00 AM, Felix Miata wrote:
http://fm.no-ip.com/Tmp/Linux/messages-gx280-esata.txt (32bit 12.1 kernel-desktop-3.1.9) ... http://fm.no-ip.com/Tmp/Linux/messages-gx620-esata.txt (32bit 12.1 kernel-desktop-3.1.9) ... Is there anything in these /var/log/messages excerpts from two different hosts that can identify which (external) sata device is connected with the reported errors? I don't see any connection between that information and the content of /dev/disk/[by-path,by-id]. Is there a way to tell if a kernel bug or a device is the problem? These errors show up in conjunction with various degrees of I/O delay, sometimes as bad as several minutes of inability to get bash to respond to keyboard input. The most recent time some commands would result in a simple segfaulted. Others would produce bash: input/output error. Ctrl-Alt-Del failed to reboot, as did init 6 (input/output error), forcing me to use power switch and wait a seeming eternity for fsck on 1TB+ partitions to complete.
Currently, no idea why the command timeout when host try to write/read device through DMA.
Did you try to use this eSATA box on other OS?
This eSATA box is a third replacement of an original (total of 4 IIRC) bought last September from Newegg: http://www.newegg.com/Product/Product.aspx?Item=N82E16817173042 This device is a v2 model purchased only after buying 5 of the v1 model and never having had any trouble with the v1 models. I've not yet had errors with this 4th one when connected to WinXP, but I did with the others. I rarely have reason to connect it to Windows, and have really only done it when problems using the device with a STB cause NTFS corruption that the STB could not fix and required CHKDSK to repair. Until today I had only connected the device to the STB, WinXP, and multiple machines running OS 11.4, 12.1 & 12.2, with various kernels on each oS. They are all multiboot hosts. Today I got Mandriva Cooker running on one machine with eSATA (in SATA mode only) available via both the motherboard ICH6 controller and PCI add-in card SiL 3512. Now I can verify the problem is not exclusive to openSUSE, but is exclusive to the v2 device, regardless which HD is installed in it. It also manifests regardless whether attached to ICH6 or SiL 3512. None of my v1 devices give this trouble.
And, Please kindly file a bug on https://bugzilla.novell.com and attached on your whole /var/log/messages, /var/log/boot.msg and lsmod result.
I will investigate it.
Doesn't this look more like upstream? Shouldn't I file there, if at all? Maybe upstream and you can combine to provide some workaround if I file there, since the manufacturer doesn't seem able to supply a usable replacement? I was expecting this to be a device problem to be corrected by the manufacturer, and hoping for information from this thread to present to the manufacturer to confirm its faulty replacement design and/or firmware, but certainly I'll file a bug it you think it's the right way to proceed. -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org