On Thu, 26 Jan 2006, Igor Yakushin wrote: > Alan, > > > > > That file doesn't contain any of the usb-storage debugging messages. > > > Hmm, I thought /var/log/messages would keep everything that happened. Or is > dmesg more detailed?
/var/log/messages keeps only the messages for which it has been configured (see /etc/syslog.conf). dmesg displays all of the kernel's messages, but only up to the limit of the internal buffer, and it doesn't display messages coming from anywhere except the kernel. > > I'd > > like to see the debugging log from a simple experiment where you try to > > copy a small amount of data and the device gets disconnected. > > OK, except that it would not get disconnected from a small amount of data. It > typically takes at > many GB of copying to break USB connection. I tried to do it yesterday but > after an hour of > running my usbburn test the computer completely froze with > ata2 : command 0x25 timeout, stat 0x50 host_stat 0x24 > messages and I had to push the power button. That's not good for debugging, because the usb-storage debugging output sometimes seems to be as large as the amount of data you're transferring! (Okay, that's an exaggeration, but it's still a tremendous amount of debugging output.) Anyway, the errors you've been getting suggest that your most pressing problem lies with the ata2 driver. > > That was with the third patch and not the second? So apparently the > > second patch isn't the only thing that messes up your computer. > > Yes, third patch (not yours). > > I tried to disable ACPI. If I give acpi=off option as a kernel parameter, the > laptop would boot > extremely slow and eventually those ata2 timeout messages appear and the boot > never completes (at > least I had no patience to wait and pushed the power button). acpid is also > running as a service. > I stopped it, ran my test and in an hour the computer froze again with ata2 > timeout messages. > Perhaps, > I should rewrite the test so that it deals only with USB and not with > internal disk to try to > isolate the problem. Yes, separating the two possible problems sources would definitely be a good thing. Alan Stern ------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642 _______________________________________________ [email protected] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-users
