<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> <html> <head> <meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type"> <title></title> </head> <body bgcolor="#ffffff" text="#000000"> I am willing to bet that you have one of those lovely drives/chipsets that reports it has finished a write but has simply cached it. A year back I upgraded from 80 gig drives to 160 gig drives and windows started crapping out left and right (the last thing windows writes to disk is the registry, lose that, lose windows). I went back to the 80 gig drives (for primary disk) and switched to linux to avoid the incredible frailty of the windows shutdown. I would try doing whatever it takes to prevent poweroff/reboot of the system on a shutdown (disable acpi, turn off caching, and do halt shutdowns, not poweroffs) and see if that fixes your problem. Then I would throw away those maxtor drives (I'm not prejudiced here am I???). Actually, I would use them in an external usb enclosure and never put them in your machine again. Matt skoal wrote: <blockquote cite="midskoal.1uwsja@gs1.ubuntuforums.org" type="cite"> <pre wrap="">Piltdownster Wrote: </pre> <blockquote type="cite"> <pre wrap="">skoal wrote: </pre> </blockquote> <pre wrap=""> </pre> <pre wrap=""> </pre> <blockquote type="cite"> <blockquote type="cite"> <pre wrap="">When shutting down (or rebooting) from Ubuntu, have you at any point </pre> </blockquote> </blockquote> <pre wrap=""> </pre> <blockquote type="cite"> <blockquote type="cite"> <pre wrap="">ever seen it hang _just_ before it reboots (or tries to)? If so, </pre> </blockquote> </blockquote> <pre wrap=""> </pre> <blockquote type="cite"> <blockquote type="cite"> <pre wrap="">although I thought this was old news by now, you could be having </pre> </blockquote> <pre wrap="">ACPI </pre> </blockquote> <pre wrap=""> </pre> <blockquote type="cite"> <blockquote type="cite"> <pre wrap="">issues with your particular hardware, even though it apparently does </pre> </blockquote> </blockquote> <pre wrap=""> </pre> <blockquote type="cite"> <blockquote type="cite"> <pre wrap="">not affect you on Windows. </pre> </blockquote> </blockquote> <pre wrap=""> </pre> <pre wrap=""> </pre> <blockquote type="cite"> <pre wrap="">I haven't noticed any odd "hanging" before reboot. Must admit, the ACPI </pre> </blockquote> <pre wrap=""> </pre> <blockquote type="cite"> <pre wrap="">issue passed me by! If it is an ACPI problem, is it possible to switch ACPI </pre> </blockquote> <pre wrap=""> </pre> <blockquote type="cite"> <pre wrap="">off by editing grub's menu.lst, i.e. </pre> </blockquote> <pre wrap=""> </pre> <pre wrap=""> </pre> <blockquote type="cite"> <pre wrap="">kernel /boot/vmlinuz-2.6.8.1-3-386 root=/dev/hda1 ro quiet splash acpi=off </pre> </blockquote> <pre wrap=""> </pre> <pre wrap=""> </pre> <blockquote type="cite"> <pre wrap="">Thanks </pre> </blockquote> <pre wrap=""> </pre> <pre wrap=""> </pre> <pre wrap=""> </pre> <blockquote type="cite"> <pre wrap="">-- </pre> </blockquote> <pre wrap=""> </pre> <blockquote type="cite"> <pre wrap="">ubuntu-users mailing list </pre> </blockquote> <pre wrap=""> </pre> <blockquote type="cite"> <pre wrap=""><a class="moz-txt-link-abbreviated" href="mailto:ubuntu-users@lists.ubuntu.com">ubuntu-users@lists.ubuntu.com</a> </pre> </blockquote> <pre wrap=""> </pre> <blockquote type="cite"> <pre wrap=""><a class="moz-txt-link-freetext" href="http://lists.ubuntu.com/mailman/listinfo/ubuntu-users">http://lists.ubuntu.com/mailman/listinfo/ubuntu-users</a> </pre> </blockquote> <pre wrap=""> Yes, sir. I believe that 'acpi=off' should work. However, I only mentioned ACPI as a potential cause, since (even as late as 2.6.10) some people had a hanging issue just before reboot and (I guess) the kernel never flushed it's I/O buffers, and consequently, on subsequent reboots people got inconsistent filesystems, and progressively worse. However, those cases were rare and I only vaguely recall hearing about them on some laptops and VIA chipsets. Whatever the case, I was just poking my guessing stick in the dark there. If you never had those issues, I doubt that's your problem. 1. If you run fsck on it and clean it up, on subsequent re-boots does it give you the _exact_ same inode errors? Different inode block? 2. Have you by chance enabled 'cache-writing' for that drive, using hdparm? What does 'sudo hdparm /dev/hdb' return (just for kicks)? Maybe you can "tune-down" your drive using hdparm (if necessary) and see if that inconsistent filesystem state returns... \\//_ </pre> </blockquote> </body> </html>