From lgj at usenix.org Wed Sep 1 12:27:49 2010 From: lgj at usenix.org (Lionel Garth Jones) Date: Wed, 1 Sep 2010 09:27:49 -0700 Subject: [Linux] LISA '10 Registration Now Open Message-ID: <10482C49-3069-4A7B-AE1B-E79FC5C509CE@usenix.org> On behalf of all the LISA '10 organizers, I'd like to invite you to join us for a career-building adventure in San Jose, CA, at the 24th Large Installation System Administration Conference, November 7-12, 2010: http://www.usenix.org/lisa10/proga For over 20 years, the Large Installation System Administration Conference (LISA) has been the must-attend system administration conference. It offers an unparalleled opportunity to meet and mingle with the leaders of the system administration industry. Take advantage of 6 days of training that offer face-to-face time with expert instructors, including: -- David N. Blank-Edelman on Over the Edge System Administration -- Patrick Ben Koetter and Ralf Hildebrandt on Dovecot and Postfix Administration We're again offering series of classes focusing on some of the most important topics you'll encounter including: -- The Virtualization Series, offering both new and repeat classesthat provide the latest virtualization information by instructors such as John Arrasjid and Richard McDougall -- New! The Linux Security and Administration Series, featuringin- depth Linux training by experts including Ted Ts'o and Rik Farrow -- New! The Super Sysadmin Series, showcasing techniques fortime and project management, raising your visibility, and other key skills to take your career to the next level by expert instructors including Tom Limoncelli and Strata Rose Chalup. The full training program can be found at http://www.usenix.org/events/lisa10/training/ In addition to the training, 3 days of technical sessions include top-notch refereed papers, Practice and Experience Reports, informative invited talks, expert Guru Is In sessions, and a poster session. http://www.usenix.org/events/lisa10/tech/ Over a dozen invited talks feature our most impressive slate of speakers to date. They include: -- Keynote Address: "The LHC Computing Challenge: Preparation, Reality and Future Outlook," by Tony Cass, CERN -- Closing Session: "Look! Up in the Sky! It's a Bird! It's a Plane! It's a Sysadmin!," by David N. Blank-Edelman, Northeastern University CCIS -- "10,000,000,000 Files Available Anywhere: NFS at Dreamworks," by Sean Kamath and Mike Cutler, PDI/Dreamworks -- "Operations at Twitter: Scaling Beyond 100 Million Users," by John Adams, Twitter LISA is the leading forum for presenting new research in system administration. This year's top-tier research showcases work covering key topics such as configuration tools and firewall analysis. The Practice and Experience Reports provide real-world examples from a variety of topics, including implementing IPv6, configuration management for Mac OS X, and more. Bring your questions to the experts in the Guru Is In sessions to unravel your greatest technical mysteries. Explore the latest commercial innovations at the Vendor Exhibition. Benefit from opportunities for peer interaction (a.k.a. the "Hallway Track"). Plus: workshops, posters, BoFs, and more Discounts are available! -- Registration: http://www.usenix.org/events/lisa10/discounts.html -- Airfare and hotel: http://www.usenix.org/events/lisa10/hotel.html Help us promote! -- Banners and buttons: http://www.usenix.org/events/lisa10/promote.html -- Facebook: http://www.usenix.org/events/lisa10/facebook -- Twitter: http://twitter.com/LISAConference For complete program information and to register, see http://www.usenix.org/lisa10/proga Early registration discounts are now available. Register by Monday, October 18, and save up to $300! We're pleased to bring LISA to San Jose, CA, and we look forward to seeing you there. On behalf of the LISA '10 Program Committee, Rudi van Drunen, Competa IT and Xlexit Technology, The Netherlands LISA '10 Program Chair lisa10chair at usenix.org ----------------------------------------------------------- LISA '10: 24th Large Installation System Administration Conference http://www.usenix.org/lisa10/proga November 7-12, 2010, San Jose, CA Early Bird Registration Deadline: October 18, 2010 Sponsored by USENIX in cooperation with LOPSA and SNIA ----------------------------------------------------------- From andy.wilkes at rocketmail.com Wed Sep 8 00:40:47 2010 From: andy.wilkes at rocketmail.com (Andy Wilkes) Date: Tue, 7 Sep 2010 21:40:47 -0700 (PDT) Subject: [Linux] how to boot from a u.s.b. cruzer Message-ID: <680550.8694.qm@web112901.mail.gq1.yahoo.com> hi, i downloaded a copy of fedora/linux to a u.s.b. cruzer, i burned the iso image, i think everything went fine, when i restarted my computer and told it to boot from the said drive,it started,the fedora page came up,it stated loading,3 bars of progress each a different color went across the bottom of my screen,then it stopped, it never got to the sign-on part of it? the u.s.b. software the fedora page said to use downloaded but a message came up saying i needed atleast 30 meg,but i only had 18 meg, could that be the problem why the cruzer didn't work and fedora load? can someone help,i'm really interested in linux, i really like it.thanks for any help. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100907/90ddb6c1/attachment.html From robert.citek at gmail.com Wed Sep 8 02:49:41 2010 From: robert.citek at gmail.com (Robert Citek) Date: Wed, 8 Sep 2010 02:49:41 -0400 Subject: [Linux] how to boot from a u.s.b. cruzer In-Reply-To: <680550.8694.qm@web112901.mail.gq1.yahoo.com> References: <680550.8694.qm@web112901.mail.gq1.yahoo.com> Message-ID: On Wed, Sep 8, 2010 at 12:40 AM, Andy Wilkes wrote: > hi, i downloaded a copy of fedora/linux to a u.s.b. cruzer, i burned the iso image, i think everything went fine, when i restarted my computer and told it to boot from the said drive,it started,the fedora page came up,it stated loading,3 bars of progress each a different color went across the bottom of my screen,then it stopped, it never got to the sign-on part of it? the u.s.b. software the fedora page said to use downloaded but a message came up saying i needed atleast 30 meg,but i only had 18 meg, could that be the problem why the cruzer didn't work and fedora load? can someone help,i'm really interested in linux, i really like it.thanks for any help. Sorry, but I'm having a hard time parsing your text. Some questions: - How big is your USB Cruzer? - How big is the ISO you downloaded from Fedora? - Are you booting the system from the CD-ROM or from the USB Cruzer? Regards, - Robert From giangiammy at gmail.com Wed Sep 8 04:31:53 2010 From: giangiammy at gmail.com (Gianluca Moro) Date: Wed, 8 Sep 2010 10:31:53 +0200 Subject: [Linux] how to boot from a u.s.b. cruzer In-Reply-To: <680550.8694.qm@web112901.mail.gq1.yahoo.com> References: <680550.8694.qm@web112901.mail.gq1.yahoo.com> Message-ID: hi, On Wed, Sep 8, 2010 at 6:40 AM, Andy Wilkes wrote: > hi, i downloaded a copy of fedora/linux to a u.s.b. cruzer, i burned the > iso I do not understand how you copied linux to usb: I usually use http://unetbootin.sourceforge.net/ from an ISO makes a bootable USB. Also check the MD% of the ISO image. Anyway, if the space check fails ... I think you must fix it! bye giammy -- Gianluca Moro http://www.cloudusb.net -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100908/f06cfd08/attachment.html From andy.wilkes at rocketmail.com Wed Sep 8 04:46:27 2010 From: andy.wilkes at rocketmail.com (Andy Wilkes) Date: Wed, 8 Sep 2010 01:46:27 -0700 (PDT) Subject: [Linux] how to boot from a u.s.b. cruzer In-Reply-To: Message-ID: <915564.15549.qm@web112907.mail.gq1.yahoo.com> hi, the cruzer is 4 gig,i'm booting it from the cruzer which it said is possible, and fedora is 691,200 kb ? --- On Wed, 9/8/10, Robert Citek wrote: From: Robert Citek Subject: Re: [Linux] how to boot from a u.s.b. cruzer To: linux at flux.org Date: Wednesday, September 8, 2010, 6:49 AM On Wed, Sep 8, 2010 at 12:40 AM, Andy Wilkes wrote: > hi, i downloaded a copy of fedora/linux to a u.s.b. cruzer, i burned the iso image, i think everything went fine, when i restarted my computer and told it to boot from the said drive,it started,the fedora page came up,it stated loading,3 bars of progress each a different color went across the bottom of my screen,then it stopped, it never got to the sign-on part of it? the u.s.b. software the fedora page said to use downloaded but a message came up saying i needed atleast 30 meg,but i only had 18 meg, could that be the problem why the cruzer didn't work and fedora load? can someone help,i'm really interested in linux, i really like it.thanks for any help. Sorry, but I'm having a hard time parsing your text.? Some questions: - How big is your USB Cruzer? - How big is the ISO you downloaded from Fedora? - Are you booting the system from the CD-ROM or from the USB Cruzer? Regards, - Robert _______________________________________________ Linux mailing list Linux at flux.org http://www.flux.org/mailman/listinfo/linux -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100908/f6914693/attachment.html From andy.wilkes at rocketmail.com Wed Sep 8 04:51:53 2010 From: andy.wilkes at rocketmail.com (Andy Wilkes) Date: Wed, 8 Sep 2010 01:51:53 -0700 (PDT) Subject: [Linux] how to boot from a u.s.b. cruzer In-Reply-To: Message-ID: <467367.89615.qm@web112914.mail.gq1.yahoo.com> hi, i downloaded the iso image to my computer, it's 691,200 kb's, i burned the iso to a 4 gig u.s.b cruzer flash drive using a program thats made to do that, it's like burning an iso to dvd burner, but just to a flash drive, then you can tell your computer to boot from the flash drive,like tewlling it to boot from the dvd rom. --- On Wed, 9/8/10, Gianluca Moro wrote: From: Gianluca Moro Subject: Re: [Linux] how to boot from a u.s.b. cruzer To: linux at flux.org Date: Wednesday, September 8, 2010, 8:31 AM hi, On Wed, Sep 8, 2010 at 6:40 AM, Andy Wilkes??wrote: hi, i downloaded a copy of fedora/linux to a u.s.b. cruzer, i burned the iso? ? I do not understand how you copied linux to usb: I usually use?http://unetbootin.sourceforge.net/ from an ISO makes a bootable USB. Also check the MD% of the ISO image. Anyway, if the space check fails ... I think you must fix it! bye? giammy -- Gianluca Moro? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?? http://www.cloudusb.net -----Inline Attachment Follows----- _______________________________________________ Linux mailing list Linux at flux.org http://www.flux.org/mailman/listinfo/linux -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100908/d5f566a2/attachment.html From adamglass at gmail.com Wed Sep 8 09:55:30 2010 From: adamglass at gmail.com (Adam Glass) Date: Wed, 8 Sep 2010 09:55:30 -0400 Subject: [Linux] how to boot from a u.s.b. cruzer In-Reply-To: <467367.89615.qm@web112914.mail.gq1.yahoo.com> References: <467367.89615.qm@web112914.mail.gq1.yahoo.com> Message-ID: Perhaps the error message is talking about hard disk space. I think some live images use hard drive for temporary storage and swap. Are you trying this on a very old computer with a very small (by today's standards) disk drive? Or to a computer whose hard drive is almost full? --Adam On Wed, Sep 8, 2010 at 4:51 AM, Andy Wilkes wrote: > hi, i downloaded the iso image to my computer, it's 691,200 kb's, i burned > the iso to a 4 gig u.s.b cruzer flash drive using a program thats made to do > that, it's like burning an iso to dvd burner, but just to a flash drive, > then you can tell your computer to boot from the flash drive,like tewlling > it to boot from the dvd rom. > > --- On *Wed, 9/8/10, Gianluca Moro * wrote: > > > From: Gianluca Moro > > Subject: Re: [Linux] how to boot from a u.s.b. cruzer > To: linux at flux.org > Date: Wednesday, September 8, 2010, 8:31 AM > > > hi, > > On Wed, Sep 8, 2010 at 6:40 AM, Andy Wilkes > > wrote: > > hi, i downloaded a copy of fedora/linux to a u.s.b. cruzer, i burned the > iso > > > I do not understand how you copied linux to usb: > I usually use http://unetbootin.sourceforge.net/ > from an ISO makes a bootable USB. > Also check the MD% of the ISO image. > Anyway, if the space check fails ... I think you must fix it! > > bye > giammy > > -- > Gianluca Moro > http://www.cloudusb.net > > > > -----Inline Attachment Follows----- > > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux > > > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100908/a1e1db2e/attachment-0001.html From dwilliams at dtw-consulting.com Wed Sep 8 22:33:18 2010 From: dwilliams at dtw-consulting.com (David Williams) Date: Wed, 08 Sep 2010 22:33:18 -0400 Subject: [Linux] Failed Hard Drive Message-ID: <4C88476E.4020700@dtw-consulting.com> Seems that one of my hard drives has decided to die on me :( I'm using Backuppc so I really hope that I can recover everything that was on this drive. When I boot up my box I see the message /dev/sdc1: recovering journal Then there are messages for mounting my other files systems then the system seems to hangs for a while, and then there are some error messages relating to /dev/sdc1 which I can't read because they scroll off too fast. Once into Linux if I try to mount /music (which is /dev/sdc1, the only partition on /dev/sdc) I get the following error message: mount: special device UUID=xxxxxxxxxxxxxxxxx does not exist If I run fsck on /dev/sdc then I get the following messages: Attempt to read block from filesystem resulted in short read while trying to open /dev/sdc Could this be a zero-length partition? I'm at a loss at what to do, if there is anything at all that I can do to try and get this drive mounted again. Any help would be much appreciated. -- ------------------------------------------------------------------------ David Williams Check out our WebOS mobile phone app for the Palm Pre and Pixi: Golf Caddie | Golf Caddie Forum | Golf Caddie FAQ by DTW-Consulting, Inc. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100908/02849920/attachment.html -------------- next part -------------- A non-text attachment was scrubbed... Name: GClogo.png Type: image/png Size: 9393 bytes Desc: not available Url : http://www.flux.org/pipermail/linux/attachments/20100908/02849920/attachment.png From dwilliams at dtw-consulting.com Thu Sep 9 08:34:50 2010 From: dwilliams at dtw-consulting.com (David Williams) Date: Thu, 09 Sep 2010 08:34:50 -0400 Subject: [Linux] Failed Hard Drive In-Reply-To: <4C88476E.4020700@dtw-consulting.com> References: <4C88476E.4020700@dtw-consulting.com> Message-ID: <4C88D46A.3070503@dtw-consulting.com> I put the drive onto another SATA controller on the motherboard and still no joy, so it's definitely looking as though it's the hard drive. If I go into the computers setup it recognises the hard drive, so that's something at least. It's a Seagate drive and under warranty (just about) so I'm going to download their tools, pop the drive in my windows machine and see what the Seagate tools report. Open to suggestions though of what I can do under Linux to try and get this puppy working again. ------------------------------------------------------------------------ David Williams Check out our WebOS mobile phone app for the Palm Pre and Pixi: Golf Caddie | Golf Caddie Forum | Golf Caddie FAQ by DTW-Consulting, Inc. On 9/8/2010 10:33 PM, David Williams wrote: > Seems that one of my hard drives has decided to die on me :( > I'm using Backuppc so I really hope that I can recover everything that > was on this drive. > > When I boot up my box I see the message > > /dev/sdc1: recovering journal > > Then there are messages for mounting my other files systems then the > system seems to hangs for a while, and then there are some error > messages relating to /dev/sdc1 which I can't read because they scroll > off too fast. > > Once into Linux if I try to mount /music (which is /dev/sdc1, the only > partition on /dev/sdc) I get the following error message: > > mount: special device UUID=xxxxxxxxxxxxxxxxx does not exist > > If I run fsck on /dev/sdc then I get the following messages: > > Attempt to read block from filesystem resulted in short read while > trying to open /dev/sdc > Could this be a zero-length partition? > > I'm at a loss at what to do, if there is anything at all that I can do > to try and get this drive mounted again. > > Any help would be much appreciated. > -- > ------------------------------------------------------------------------ > > David Williams > Check out our WebOS mobile phone app for the Palm Pre and Pixi: > Golf Caddie > | Golf Caddie Forum > | Golf Caddie FAQ > by DTW-Consulting, > Inc. > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100909/1c4a513d/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: image/png Size: 9393 bytes Desc: not available Url : http://www.flux.org/pipermail/linux/attachments/20100909/1c4a513d/attachment-0001.png From stan at gate.net Thu Sep 9 08:36:52 2010 From: stan at gate.net (Stan Fields) Date: Thu, 9 Sep 2010 08:36:52 -0400 Subject: [Linux] Failed Hard Drive In-Reply-To: <4C88476E.4020700@dtw-consulting.com> References: <4C88476E.4020700@dtw-consulting.com> Message-ID: <20100909083652.70cc0af2@stanltop.questmktg.com> On Wed, 08 Sep 2010 22:33:18 -0400 David Williams wrote: > Seems that one of my hard drives has decided to die on me :( > I'm using Backuppc so I really hope that I can recover everything > that was on this drive. > > When I boot up my box I see the message > > /dev/sdc1: recovering journal > > Then there are messages for mounting my other files systems then the > system seems to hangs for a while, and then there are some error > messages relating to /dev/sdc1 which I can't read because they scroll > off too fast. > > Once into Linux if I try to mount /music (which is /dev/sdc1, the > only partition on /dev/sdc) I get the following error message: > > mount: special device UUID=xxxxxxxxxxxxxxxxx does not exist > > If I run fsck on /dev/sdc then I get the following messages: > > Attempt to read block from filesystem resulted in short read while > trying to open /dev/sdc > Could this be a zero-length partition? > > I'm at a loss at what to do, if there is anything at all that I can > do to try and get this drive mounted again. > > Any help would be much appreciated. You can try dd-rescue to another disk of equal or larger size or I think the distro "RIP (recovery is possible) will help you -- Stan The Computer Man aka: Stan Fields Everywhere is walking distance if you have the time. From brianshoran at gmail.com Thu Sep 9 08:47:23 2010 From: brianshoran at gmail.com (Brian Horan) Date: Thu, 9 Sep 2010 08:47:23 -0400 Subject: [Linux] Failed Hard Drive In-Reply-To: References: <4C88476E.4020700@dtw-consulting.com> <20100909083652.70cc0af2@stanltop.questmktg.com> Message-ID: Did you try fsck-ing sdc or sdc1? Also, i've had some luck in the past with putting failed disks in the freezer for a while, taking them out, and copying the data elsewhere. Yes, it sounds odd...but it has worked-not for filesystem issues, but for disk issues. On Sep 9, 2010 8:35 AM, "Stan Fields" wrote: On Wed, 08 Sep 2010 22:33:18 -0400 David Williams wrote: > Seems t... You can try dd-rescue to another disk of equal or larger size or I think the distro "RIP (recovery is possible) will help you -- Stan The Computer Man aka: Stan Fields Everywhere is walking distance if you have the time. _______________________________________________ Linux mailing list Linux at flux.org http://www.flux.org/mailman/listinfo/linux -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100909/41027193/attachment.html From dwilliams at dtw-consulting.com Thu Sep 9 08:50:36 2010 From: dwilliams at dtw-consulting.com (David Williams) Date: Thu, 09 Sep 2010 08:50:36 -0400 Subject: [Linux] Failed Hard Drive In-Reply-To: References: <4C88476E.4020700@dtw-consulting.com> <20100909083652.70cc0af2@stanltop.questmktg.com> Message-ID: <4C88D81C.6010409@dtw-consulting.com> Yes, I tried to run fsck on /dev/sdc and /dev/sdc1. I put that in my original post. Didn't work. Might try the freezer thing then....worth a shot I guess. ------------------------------------------------------------------------ David Williams Check out our WebOS mobile phone app for the Palm Pre and Pixi: Golf Caddie | Golf Caddie Forum | Golf Caddie FAQ by DTW-Consulting, Inc. On 9/9/2010 8:47 AM, Brian Horan wrote: > > Did you try fsck-ing sdc or sdc1? Also, i've had some luck in the past > with putting failed disks in the freezer for a while, taking them out, > and copying the data elsewhere. Yes, it sounds odd...but it has > worked-not for filesystem issues, but for disk issues. > >> On Sep 9, 2010 8:35 AM, "Stan Fields" > > wrote: >> >> On Wed, 08 Sep 2010 22:33:18 -0400 >> David Williams > > wrote: >> >> > Seems t... >> >> You can try dd-rescue to another disk of equal or larger size or >> I think the distro "RIP (recovery is possible) will help you >> >> -- >> Stan The Computer Man >> aka: Stan Fields >> >> Everywhere is walking distance if you have the time. >> _______________________________________________ >> Linux mailing list >> Linux at flux.org >> http://www.flux.org/mailman/listinfo/linux > > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100909/7f7f1ea6/attachment.html -------------- next part -------------- A non-text attachment was scrubbed... Name: GClogo.png Type: image/png Size: 9393 bytes Desc: not available Url : http://www.flux.org/pipermail/linux/attachments/20100909/7f7f1ea6/attachment.png From joee at lithodyne.net Thu Sep 9 08:58:19 2010 From: joee at lithodyne.net (Joseph Ellis) Date: Thu, 9 Sep 2010 08:58:19 -0400 Subject: [Linux] Failed Hard Drive In-Reply-To: References: <4C88476E.4020700@dtw-consulting.com> <20100909083652.70cc0af2@stanltop.questmktg.com> Message-ID: <4817D2E6-5E09-4EE9-A56E-FA8E0F23596F@lithodyne.net> Interesting. I'll have to try that. I have a failed disk that I've been holding onto so hopefully I'll be able to finally recover the data. Sent from my iPhone On Sep 9, 2010, at 8:47 AM, Brian Horan wrote: > Did you try fsck-ing sdc or sdc1? Also, i've had some luck in the > past with putting failed disks in the freezer for a while, taking > them out, and copying the data elsewhere. Yes, it sounds odd...but > it has worked-not for filesystem issues, but for disk issues. > >> On Sep 9, 2010 8:35 AM, "Stan Fields" wrote: >> >> On Wed, 08 Sep 2010 22:33:18 -0400 >> David Williams wrote: >> >> > Seems t... >> >> You can try dd-rescue to another disk of equal or larger size or >> I think the distro "RIP (recovery is possible) will help you >> >> -- >> Stan The Computer Man >> aka: Stan Fields >> >> Everywhere is walking distance if you have the time. >> _______________________________________________ >> Linux mailing list >> Linux at flux.org >> http://www.flux.org/mailman/listinfo/linux > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100909/d98651f8/attachment-0001.html From brianshoran at gmail.com Thu Sep 9 09:32:34 2010 From: brianshoran at gmail.com (Brian Horan) Date: Thu, 9 Sep 2010 09:32:34 -0400 Subject: [Linux] Failed Hard Drive In-Reply-To: <4817D2E6-5E09-4EE9-A56E-FA8E0F23596F@lithodyne.net> References: <4C88476E.4020700@dtw-consulting.com> <20100909083652.70cc0af2@stanltop.questmktg.com> <4817D2E6-5E09-4EE9-A56E-FA8E0F23596F@lithodyne.net> Message-ID: i've attempted it about 30 times over the years, and have been successful with it maybe 22-23 times.. you don't get tons of time - it seems that once the disk warms up, it's back to its previous state of fucture...pardon my language... On Thu, Sep 9, 2010 at 8:58 AM, Joseph Ellis wrote: > Interesting. I'll have to try that. I have a failed disk that I've been > holding onto so hopefully I'll be able to finally recover the data. > > Sent from my iPhone > > On Sep 9, 2010, at 8:47 AM, Brian Horan wrote: > > Did you try fsck-ing sdc or sdc1? Also, i've had some luck in the past with > putting failed disks in the freezer for a while, taking them out, and > copying the data elsewhere. Yes, it sounds odd...but it has worked-not for > filesystem issues, but for disk issues. > > On Sep 9, 2010 8:35 AM, "Stan Fields" < stan at gate.net> > wrote: > > On Wed, 08 Sep 2010 22:33:18 -0400 > David Williams < > dwilliams at dtw-consulting.com> wrote: > > > Seems t... > You can try dd-rescue to another disk of equal or larger size or > I think the distro "RIP (recovery is possible) will help you > > -- > Stan The Computer Man > aka: Stan Fields > > Everywhere is walking distance if you have the time. > _______________________________________________ > Linux mailing list > Linux at flux.org > > http://www.flux.org/mailman/listinfo/linux > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux > > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100909/9068566c/attachment.html From info at carlc.com Thu Sep 9 09:53:22 2010 From: info at carlc.com (CarlC - Information) Date: Thu, 9 Sep 2010 09:53:22 -0400 Subject: [Linux] Failed Hard Drive In-Reply-To: References: Message-ID: > -----Original Message----- > Date: Thu, 09 Sep 2010 08:50:36 -0400 > From: David Williams > Yes, I tried to run fsck on /dev/sdc and /dev/sdc1. I put > that in my original post. Didn't work. > Might try the freezer thing then....worth a shot I guess. The real trick is to first use smartd/smartctl. Using smartctl, you can see if the disk is truly gone: Smartctl -a /dev/sdc | more Look at the ECC and Raw_Read_Error_Rate. Also, the Seek_Error_Rate will show if your taking seek errors as well. If Adam ever wants, I can give an entire presentation on Disk drive monitoring and health. Smartd/smartctl is your saviour :) . Especially when you can use smartd to monitor your drives, test them when you want, and even test individual drives in a Raid set (provided the Raid controller allows it). I have yet to lose a drive without smartd warning me, usually weeks to months in advanced. For my Hosting business, it's saved me so much time as it's one of the best ways to be in preventative maintenance mode. As to recovery, between and ice pack/heating pad, and luck, well, now we know why backups are critical. Professional shops can take the drive appart and use tricks like off-center head readings to get the data back, but then it's all a matter of how much money is the data worth. Carl P.s. if your drive is USB, smartd may not work. You may have to take the drive out of the USB containter and directly connect it. I've even seen a case where the USB controller died, but the drive was fine and we could easily recover the data while the drive was in direct connection. From brianshoran at gmail.com Thu Sep 9 09:57:03 2010 From: brianshoran at gmail.com (Brian Horan) Date: Thu, 9 Sep 2010 09:57:03 -0400 Subject: [Linux] Failed Hard Drive In-Reply-To: References: Message-ID: I absolutely agree with Carl. On Thu, Sep 9, 2010 at 9:53 AM, CarlC - Information wrote: > > > > -----Original Message----- > > Date: Thu, 09 Sep 2010 08:50:36 -0400 > > From: David Williams > > Yes, I tried to run fsck on /dev/sdc and /dev/sdc1. I put > > that in my original post. Didn't work. > > Might try the freezer thing then....worth a shot I guess. > > The real trick is to first use smartd/smartctl. Using smartctl, you can see > if the disk is truly gone: > > Smartctl -a /dev/sdc | more > > Look at the ECC and Raw_Read_Error_Rate. Also, the Seek_Error_Rate will > show if your taking seek errors as well. > > If Adam ever wants, I can give an entire presentation on Disk drive > monitoring and health. Smartd/smartctl is your saviour :) . > Especially when you can use smartd to monitor your drives, test them when > you want, and even test individual drives in a Raid set > (provided the Raid controller allows it). > > I have yet to lose a drive without smartd warning me, usually weeks to > months in advanced. For my Hosting business, it's saved me > so much time as it's one of the best ways to be in preventative maintenance > mode. > > As to recovery, between and ice pack/heating pad, and luck, well, now we > know why backups are critical. Professional shops can > take the drive appart and use tricks like off-center head readings to get > the data back, but then it's all a matter of how much > money is the data worth. > > Carl > > P.s. if your drive is USB, smartd may not work. You may have to take the > drive out of the USB containter and directly connect it. > I've even seen a case where the USB controller died, but the drive was fine > and we could easily recover the data while the drive > was in direct connection. > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100909/b6c90dde/attachment.html From dwilliams at dtw-consulting.com Thu Sep 9 10:07:06 2010 From: dwilliams at dtw-consulting.com (David Williams) Date: Thu, 09 Sep 2010 10:07:06 -0400 Subject: [Linux] Failed Hard Drive In-Reply-To: References: Message-ID: <4C88EA0A.4000900@dtw-consulting.com> Carl, I will try the smartctl -a /dev/sdc command and see what comes out of that. I'd be interested to learn more about setting up smart monitor on my drives. Nothing that I have is that critical, this is just my home box, and I do back up most of what I deem is important to me. The drive that went bad has all my music on it in FLAC format, which took hours and hours to rip. I'd hate to have to go through all that again :-( Will pop the drive back in my Linux box and try the smartctl command out. Thanks. ------------------------------------------------------------------------ David Williams Check out our WebOS mobile phone app for the Palm Pre and Pixi: Golf Caddie | Golf Caddie Forum | Golf Caddie FAQ by DTW-Consulting, Inc. On 9/9/2010 9:53 AM, CarlC - Information wrote: > > >> -----Original Message----- >> Date: Thu, 09 Sep 2010 08:50:36 -0400 >> From: David Williams >> Yes, I tried to run fsck on /dev/sdc and /dev/sdc1. I put >> that in my original post. Didn't work. >> Might try the freezer thing then....worth a shot I guess. > The real trick is to first use smartd/smartctl. Using smartctl, you can see if the disk is truly gone: > > Smartctl -a /dev/sdc | more > > Look at the ECC and Raw_Read_Error_Rate. Also, the Seek_Error_Rate will show if your taking seek errors as well. > > If Adam ever wants, I can give an entire presentation on Disk drive monitoring and health. Smartd/smartctl is your saviour :) . > Especially when you can use smartd to monitor your drives, test them when you want, and even test individual drives in a Raid set > (provided the Raid controller allows it). > > I have yet to lose a drive without smartd warning me, usually weeks to months in advanced. For my Hosting business, it's saved me > so much time as it's one of the best ways to be in preventative maintenance mode. > > As to recovery, between and ice pack/heating pad, and luck, well, now we know why backups are critical. Professional shops can > take the drive appart and use tricks like off-center head readings to get the data back, but then it's all a matter of how much > money is the data worth. > > Carl > > P.s. if your drive is USB, smartd may not work. You may have to take the drive out of the USB containter and directly connect it. > I've even seen a case where the USB controller died, but the drive was fine and we could easily recover the data while the drive > was in direct connection. > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100909/d94e5d68/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: GClogo.png Type: image/png Size: 9393 bytes Desc: not available Url : http://www.flux.org/pipermail/linux/attachments/20100909/d94e5d68/attachment-0001.png From adamglass at gmail.com Thu Sep 9 12:12:30 2010 From: adamglass at gmail.com (Adam Glass) Date: Thu, 9 Sep 2010 12:12:30 -0400 Subject: [Linux] Failed Hard Drive In-Reply-To: References: Message-ID: Adam _always_ wants people to volunteer to do FLUX presentations :-) We'll talk off-list! --Adam On Thu, Sep 9, 2010 at 9:53 AM, CarlC - Information wrote: > > > > -----Original Message----- > > Date: Thu, 09 Sep 2010 08:50:36 -0400 > > From: David Williams > > Yes, I tried to run fsck on /dev/sdc and /dev/sdc1. I put > > that in my original post. Didn't work. > > Might try the freezer thing then....worth a shot I guess. > > The real trick is to first use smartd/smartctl. Using smartctl, you can see > if the disk is truly gone: > > Smartctl -a /dev/sdc | more > > Look at the ECC and Raw_Read_Error_Rate. Also, the Seek_Error_Rate will > show if your taking seek errors as well. > > If Adam ever wants, I can give an entire presentation on Disk drive > monitoring and health. Smartd/smartctl is your saviour :) . > Especially when you can use smartd to monitor your drives, test them when > you want, and even test individual drives in a Raid set > (provided the Raid controller allows it). > > I have yet to lose a drive without smartd warning me, usually weeks to > months in advanced. For my Hosting business, it's saved me > so much time as it's one of the best ways to be in preventative maintenance > mode. > > As to recovery, between and ice pack/heating pad, and luck, well, now we > know why backups are critical. Professional shops can > take the drive appart and use tricks like off-center head readings to get > the data back, but then it's all a matter of how much > money is the data worth. > > Carl > > P.s. if your drive is USB, smartd may not work. You may have to take the > drive out of the USB containter and directly connect it. > I've even seen a case where the USB controller died, but the drive was fine > and we could easily recover the data while the drive > was in direct connection. > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100909/2d60e12d/attachment.html From ckosloff at hotmail.com Thu Sep 9 16:23:07 2010 From: ckosloff at hotmail.com (Carlos Kosloff) Date: Thu, 9 Sep 2010 20:23:07 +0000 Subject: [Linux] Presentation on disk drives monitoring In-Reply-To: References: Message-ID: > If Adam ever wants, I can give an entire presentation on Disk drive monitoring and health. Smartd/smartctl is your saviour :) . Aye, I vote for it. I am still beating my forehead against the wall for having missed Jeffrey Fortin's presentation on writing device drivers. Disk recovery and computer forensics is also a fascinating subject. There is a firm in Hollywood, FL that manufactures devices for backup, raid, mirroring, etc. I am pretty sure that they would love to present their products alongside Carl's presentation, or maybe on another date. What sayst thou, Adam? Carlos Kosloff -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100909/9325b249/attachment.html From adamglass at gmail.com Thu Sep 9 16:34:00 2010 From: adamglass at gmail.com (Adam Glass) Date: Thu, 9 Sep 2010 16:34:00 -0400 Subject: [Linux] Presentation on disk drives monitoring In-Reply-To: References: Message-ID: We always need presenters, and I'll be talking to Carl about this off-list. Since we hope to have a lot of newbies attend the meeting after Software Freedom Day, this topic might be too advanced to cover right away, but definitely soon. In general we do not invite companies to present their products. When someone from a company approaches us about that, we ask them to send an engineer along with the sales person, to give a talk about something in the same field that anyone can learn from, even if they do not purchase the company's product. Then they can do a short segment on their product at the end. --Adam On Thu, Sep 9, 2010 at 4:23 PM, Carlos Kosloff wrote: > > If Adam ever wants, I can give an entire presentation on Disk drive > monitoring and health. Smartd/smartctl is your saviour :) . > > Aye, I vote for it. > I am still beating my forehead against the wall for having missed Jeffrey > Fortin's presentation on writing device drivers. > Disk recovery and computer forensics is also a fascinating subject. > There is a firm in Hollywood, FL that manufactures devices for backup, > raid, mirroring, etc. > I am pretty sure that they would love to present their products alongside > Carl's presentation, or maybe on another date. > What sayst thou, Adam? > > > Carlos Kosloff > > > > > > > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100909/5a0ad66a/attachment.html From dwilliams at dtw-consulting.com Thu Sep 9 18:51:39 2010 From: dwilliams at dtw-consulting.com (David Williams) Date: Thu, 09 Sep 2010 18:51:39 -0400 Subject: [Linux] Failed Hard Drive In-Reply-To: References: Message-ID: <4C8964FB.5060405@dtw-consulting.com> This is what I got when I tried to run smartctl on /dev/sdc smartctl -a /dev/sdc1 | more smartctl version 5.38 [i586-mandriva-linux-gnu] Copyright (C) 2002-8 Bruce Allen Home page is http://smartmontools.sourceforge.net/ Short INQUIRY response, skip product id A mandatory SMART command failed: exiting. To continue, add one or more '-T permissive' options. ------------------------------------------------------------------------ David Williams Check out our WebOS mobile phone app for the Palm Pre and Pixi: Golf Caddie | Golf Caddie Forum | Golf Caddie FAQ by DTW-Consulting, Inc. On 9/9/2010 9:53 AM, CarlC - Information wrote: > > >> -----Original Message----- >> Date: Thu, 09 Sep 2010 08:50:36 -0400 >> From: David Williams >> Yes, I tried to run fsck on /dev/sdc and /dev/sdc1. I put >> that in my original post. Didn't work. >> Might try the freezer thing then....worth a shot I guess. > The real trick is to first use smartd/smartctl. Using smartctl, you can see if the disk is truly gone: > > Smartctl -a /dev/sdc | more > > Look at the ECC and Raw_Read_Error_Rate. Also, the Seek_Error_Rate will show if your taking seek errors as well. > > If Adam ever wants, I can give an entire presentation on Disk drive monitoring and health. Smartd/smartctl is your saviour :) . > Especially when you can use smartd to monitor your drives, test them when you want, and even test individual drives in a Raid set > (provided the Raid controller allows it). > > I have yet to lose a drive without smartd warning me, usually weeks to months in advanced. For my Hosting business, it's saved me > so much time as it's one of the best ways to be in preventative maintenance mode. > > As to recovery, between and ice pack/heating pad, and luck, well, now we know why backups are critical. Professional shops can > take the drive appart and use tricks like off-center head readings to get the data back, but then it's all a matter of how much > money is the data worth. > > Carl > > P.s. if your drive is USB, smartd may not work. You may have to take the drive out of the USB containter and directly connect it. > I've even seen a case where the USB controller died, but the drive was fine and we could easily recover the data while the drive > was in direct connection. > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100909/b9c6c6de/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: GClogo.png Type: image/png Size: 9393 bytes Desc: not available Url : http://www.flux.org/pipermail/linux/attachments/20100909/b9c6c6de/attachment-0001.png From mmarschall at yahoo.com Thu Sep 9 16:41:48 2010 From: mmarschall at yahoo.com (Mike Marschall) Date: Thu, 9 Sep 2010 13:41:48 -0700 Subject: [Linux] Presentation on disk drives monitoring In-Reply-To: References: Message-ID: <206E5C2C-1323-45EE-88BE-7E92E79376C0@yahoo.com> Speaking of which, I am going to schedule my trip to FL this weekend. Adam I will work with you to schedule a talk on Massh. I would also be up to doing a talk (or maybe some smaller get togethers) about Monit, Puppet (general configuration management) and or strategies for massive scale. It will be early/mid Oct or early/mid Nov. mm On Sep 9, 2010, at 1:34 PM, Adam Glass wrote: > We always need presenters, and I'll be talking to Carl about this off-list. Since we hope to have a lot of newbies attend the meeting after Software Freedom Day, this topic might be too advanced to cover right away, but definitely soon. > > In general we do not invite companies to present their products. When someone from a company approaches us about that, we ask them to send an engineer along with the sales person, to give a talk about something in the same field that anyone can learn from, even if they do not purchase the company's product. Then they can do a short segment on their product at the end. > > > --Adam > > On Thu, Sep 9, 2010 at 4:23 PM, Carlos Kosloff wrote: > > If Adam ever wants, I can give an entire presentation on Disk drive monitoring and health. Smartd/smartctl is your saviour :) . > > Aye, I vote for it. > I am still beating my forehead against the wall for having missed Jeffrey Fortin's presentation on writing device drivers. > Disk recovery and computer forensics is also a fascinating subject. > There is a firm in Hollywood, FL that manufactures devices for backup, raid, mirroring, etc. > I am pretty sure that they would love to present their products alongside Carl's presentation, or maybe on another date. > What sayst thou, Adam? > > > Carlos Kosloff > > > > > > > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux > > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux From wordz2u at gmail.com Thu Sep 9 21:23:58 2010 From: wordz2u at gmail.com (Steve G.) Date: Thu, 9 Sep 2010 19:23:58 -0600 Subject: [Linux] Failed Hard Drive In-Reply-To: <4C8964FB.5060405@dtw-consulting.com> References: <4C8964FB.5060405@dtw-consulting.com> Message-ID: I think by now you are coming to the sad realization that your drive if fried beyond repair, right? I am not an expert on drives, but I have never recovered from short reads. I think it says that there is something wrong with the table that contains the information about the data in the drive, which is corrupted. My experience is that usually if I can't mount the drive, or run fsck on it, the game is over. Luckily for me, I rarely have any data worth saving, so I never bothered to freeze or send the server to experts. Still, you have nothing to lose by burning the vendor disk and letting it tell you your disk is gone. On a related matter, the monthly lecture make me sorry not to be living in Florida any more. That and TigerDirect. Z. On Thu, Sep 9, 2010 at 4:51 PM, David Williams wrote: > This is what I got when I tried to run smartctl on /dev/sdc > > smartctl -a /dev/sdc1 | more > smartctl version 5.38 [i586-mandriva-linux-gnu] Copyright (C) 2002-8 Bruce > Allen > Home page is http://smartmontools.sourceforge.net/ > > Short INQUIRY response, skip product id > A mandatory SMART command failed: exiting. To continue, add one or more '-T > permissive' options. > ------------------------------ > > David Williams > Check out our WebOS mobile phone app for the Palm Pre and Pixi: > Golf Caddie| Golf > Caddie Forum | Golf > Caddie FAQ by > DTW-Consulting, Inc. > > > > On 9/9/2010 9:53 AM, CarlC - Information wrote: > > > -----Original Message----- > Date: Thu, 09 Sep 2010 08:50:36 -0400 > From: David Williams > Yes, I tried to run fsck on /dev/sdc and /dev/sdc1. I put > that in my original post. Didn't work. > Might try the freezer thing then....worth a shot I guess. > > The real trick is to first use smartd/smartctl. Using smartctl, you can see if the disk is truly gone: > > Smartctl -a /dev/sdc | more > > Look at the ECC and Raw_Read_Error_Rate. Also, the Seek_Error_Rate will show if your taking seek errors as well. > > If Adam ever wants, I can give an entire presentation on Disk drive monitoring and health. Smartd/smartctl is your saviour :) . > Especially when you can use smartd to monitor your drives, test them when you want, and even test individual drives in a Raid set > (provided the Raid controller allows it). > > I have yet to lose a drive without smartd warning me, usually weeks to months in advanced. For my Hosting business, it's saved me > so much time as it's one of the best ways to be in preventative maintenance mode. > > As to recovery, between and ice pack/heating pad, and luck, well, now we know why backups are critical. Professional shops can > take the drive appart and use tricks like off-center head readings to get the data back, but then it's all a matter of how much > money is the data worth. > > Carl > > P.s. if your drive is USB, smartd may not work. You may have to take the drive out of the USB containter and directly connect it. > I've even seen a case where the USB controller died, but the drive was fine and we could easily recover the data while the drive > was in direct connection. > > _______________________________________________ > Linux mailing listLinux at flux.orghttp://www.flux.org/mailman/listinfo/linux > > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux > > -- Check out my web site - www.words2u.net -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100909/e729dfc8/attachment-0002.html -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: image/png Size: 9393 bytes Desc: not available Url : http://www.flux.org/pipermail/linux/attachments/20100909/e729dfc8/attachment-0002.png From wordz2u at gmail.com Thu Sep 9 21:23:58 2010 From: wordz2u at gmail.com (Steve G.) Date: Thu, 9 Sep 2010 19:23:58 -0600 Subject: [Linux] Failed Hard Drive In-Reply-To: <4C8964FB.5060405@dtw-consulting.com> References: <4C8964FB.5060405@dtw-consulting.com> Message-ID: I think by now you are coming to the sad realization that your drive if fried beyond repair, right? I am not an expert on drives, but I have never recovered from short reads. I think it says that there is something wrong with the table that contains the information about the data in the drive, which is corrupted. My experience is that usually if I can't mount the drive, or run fsck on it, the game is over. Luckily for me, I rarely have any data worth saving, so I never bothered to freeze or send the server to experts. Still, you have nothing to lose by burning the vendor disk and letting it tell you your disk is gone. On a related matter, the monthly lecture make me sorry not to be living in Florida any more. That and TigerDirect. Z. On Thu, Sep 9, 2010 at 4:51 PM, David Williams wrote: > This is what I got when I tried to run smartctl on /dev/sdc > > smartctl -a /dev/sdc1 | more > smartctl version 5.38 [i586-mandriva-linux-gnu] Copyright (C) 2002-8 Bruce > Allen > Home page is http://smartmontools.sourceforge.net/ > > Short INQUIRY response, skip product id > A mandatory SMART command failed: exiting. To continue, add one or more '-T > permissive' options. > ------------------------------ > > David Williams > Check out our WebOS mobile phone app for the Palm Pre and Pixi: > Golf Caddie| Golf > Caddie Forum | Golf > Caddie FAQ by > DTW-Consulting, Inc. > > > > On 9/9/2010 9:53 AM, CarlC - Information wrote: > > > -----Original Message----- > Date: Thu, 09 Sep 2010 08:50:36 -0400 > From: David Williams > Yes, I tried to run fsck on /dev/sdc and /dev/sdc1. I put > that in my original post. Didn't work. > Might try the freezer thing then....worth a shot I guess. > > The real trick is to first use smartd/smartctl. Using smartctl, you can see if the disk is truly gone: > > Smartctl -a /dev/sdc | more > > Look at the ECC and Raw_Read_Error_Rate. Also, the Seek_Error_Rate will show if your taking seek errors as well. > > If Adam ever wants, I can give an entire presentation on Disk drive monitoring and health. Smartd/smartctl is your saviour :) . > Especially when you can use smartd to monitor your drives, test them when you want, and even test individual drives in a Raid set > (provided the Raid controller allows it). > > I have yet to lose a drive without smartd warning me, usually weeks to months in advanced. For my Hosting business, it's saved me > so much time as it's one of the best ways to be in preventative maintenance mode. > > As to recovery, between and ice pack/heating pad, and luck, well, now we know why backups are critical. Professional shops can > take the drive appart and use tricks like off-center head readings to get the data back, but then it's all a matter of how much > money is the data worth. > > Carl > > P.s. if your drive is USB, smartd may not work. You may have to take the drive out of the USB containter and directly connect it. > I've even seen a case where the USB controller died, but the drive was fine and we could easily recover the data while the drive > was in direct connection. > > _______________________________________________ > Linux mailing listLinux at flux.orghttp://www.flux.org/mailman/listinfo/linux > > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux > > -- Check out my web site - www.words2u.net -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100909/e729dfc8/attachment-0003.html -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: image/png Size: 9393 bytes Desc: not available Url : http://www.flux.org/pipermail/linux/attachments/20100909/e729dfc8/attachment-0003.png From dwilliams at dtw-consulting.com Thu Sep 9 21:56:36 2010 From: dwilliams at dtw-consulting.com (David Williams) Date: Thu, 09 Sep 2010 21:56:36 -0400 Subject: [Linux] Failed Hard Drive In-Reply-To: References: <4C8964FB.5060405@dtw-consulting.com> Message-ID: <4C899054.5000607@dtw-consulting.com> Yes, pretty sure disk is gone...hasta la nunca. Downloaded the Seagate tools and it wouldn't even locate the drive, even it shows in the BIOS. Oh well, time to get a replacement drive since I am still in warranty by 10 days! ------------------------------------------------------------------------ David Williams Check out our WebOS mobile phone app for the Palm Pre and Pixi: Golf Caddie | Golf Caddie Forum | Golf Caddie FAQ by DTW-Consulting, Inc. On 9/9/2010 9:23 PM, Steve G. wrote: > I think by now you are coming to the sad realization that your drive > if fried beyond repair, right? > > I am not an expert on drives, but I have never recovered from short > reads. I think it says that there is something wrong with the table > that contains the information about the data in the drive, which is > corrupted. My experience is that usually if I can't mount the drive, > or run fsck on it, the game is over. Luckily for me, I rarely have any > data worth saving, so I never bothered to freeze or send the server to > experts. Still, you have nothing to lose by burning the vendor disk > and letting it tell you your disk is gone. > > On a related matter, the monthly lecture make me sorry not to be > living in Florida any more. That and TigerDirect. > > Z. > > > > On Thu, Sep 9, 2010 at 4:51 PM, David Williams > > > wrote: > > This is what I got when I tried to run smartctl on /dev/sdc > > smartctl -a /dev/sdc1 | more > smartctl version 5.38 [i586-mandriva-linux-gnu] Copyright (C) > 2002-8 Bruce Allen > Home page is http://smartmontools.sourceforge.net/ > > Short INQUIRY response, skip product id > A mandatory SMART command failed: exiting. To continue, add one or > more '-T permissive' options. > ------------------------------------------------------------------------ > > David Williams > Check out our WebOS mobile phone app for the Palm Pre and Pixi: > Golf Caddie > | Golf Caddie Forum > | Golf Caddie FAQ > by > DTW-Consulting, Inc. > > > > On 9/9/2010 9:53 AM, CarlC - Information wrote: >> >> >>> -----Original Message----- >>> Date: Thu, 09 Sep 2010 08:50:36 -0400 >>> From: David Williams >>> Yes, I tried to run fsck on /dev/sdc and /dev/sdc1. I put >>> that in my original post. Didn't work. >>> Might try the freezer thing then....worth a shot I guess. >> The real trick is to first use smartd/smartctl. Using smartctl, you can see if the disk is truly gone: >> >> Smartctl -a /dev/sdc | more >> >> Look at the ECC and Raw_Read_Error_Rate. Also, the Seek_Error_Rate will show if your taking seek errors as well. >> >> If Adam ever wants, I can give an entire presentation on Disk drive monitoring and health. Smartd/smartctl is your saviour :) . >> Especially when you can use smartd to monitor your drives, test them when you want, and even test individual drives in a Raid set >> (provided the Raid controller allows it). >> >> I have yet to lose a drive without smartd warning me, usually weeks to months in advanced. For my Hosting business, it's saved me >> so much time as it's one of the best ways to be in preventative maintenance mode. >> >> As to recovery, between and ice pack/heating pad, and luck, well, now we know why backups are critical. Professional shops can >> take the drive appart and use tricks like off-center head readings to get the data back, but then it's all a matter of how much >> money is the data worth. >> >> Carl >> >> P.s. if your drive is USB, smartd may not work. You may have to take the drive out of the USB containter and directly connect it. >> I've even seen a case where the USB controller died, but the drive was fine and we could easily recover the data while the drive >> was in direct connection. >> >> _______________________________________________ >> Linux mailing list >> Linux at flux.org >> http://www.flux.org/mailman/listinfo/linux > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux > > > > > -- > Check out my web site - www.words2u.net > > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100909/8dcc741e/attachment-0001.html -------------- next part -------------- A non-text attachment was scrubbed... Name: GClogo.png Type: image/png Size: 9393 bytes Desc: not available Url : http://www.flux.org/pipermail/linux/attachments/20100909/8dcc741e/attachment-0002.png -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: image/png Size: 9393 bytes Desc: not available Url : http://www.flux.org/pipermail/linux/attachments/20100909/8dcc741e/attachment-0003.png From dwilliams at dtw-consulting.com Fri Sep 10 20:39:13 2010 From: dwilliams at dtw-consulting.com (David Williams) Date: Fri, 10 Sep 2010 20:39:13 -0400 Subject: [Linux] Failed Hard Drive In-Reply-To: References: Message-ID: <4C8ACFB1.6040407@dtw-consulting.com> So, Replacement hard drive is on it's way :) and I also went out and bought a 2TB drive today from BB. Restoring all my music now, so good to know that backuppc works like a charm :) ------------------------------------------------------------------------ David Williams Check out our WebOS mobile phone app for the Palm Pre and Pixi: Golf Caddie | Golf Caddie Forum | Golf Caddie FAQ by DTW-Consulting, Inc. On 9/9/2010 9:53 AM, CarlC - Information wrote: > > >> -----Original Message----- >> Date: Thu, 09 Sep 2010 08:50:36 -0400 >> From: David Williams >> Yes, I tried to run fsck on /dev/sdc and /dev/sdc1. I put >> that in my original post. Didn't work. >> Might try the freezer thing then....worth a shot I guess. > The real trick is to first use smartd/smartctl. Using smartctl, you can see if the disk is truly gone: > > Smartctl -a /dev/sdc | more > > Look at the ECC and Raw_Read_Error_Rate. Also, the Seek_Error_Rate will show if your taking seek errors as well. > > If Adam ever wants, I can give an entire presentation on Disk drive monitoring and health. Smartd/smartctl is your saviour :) . > Especially when you can use smartd to monitor your drives, test them when you want, and even test individual drives in a Raid set > (provided the Raid controller allows it). > > I have yet to lose a drive without smartd warning me, usually weeks to months in advanced. For my Hosting business, it's saved me > so much time as it's one of the best ways to be in preventative maintenance mode. > > As to recovery, between and ice pack/heating pad, and luck, well, now we know why backups are critical. Professional shops can > take the drive appart and use tricks like off-center head readings to get the data back, but then it's all a matter of how much > money is the data worth. > > Carl > > P.s. if your drive is USB, smartd may not work. You may have to take the drive out of the USB containter and directly connect it. > I've even seen a case where the USB controller died, but the drive was fine and we could easily recover the data while the drive > was in direct connection. > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100910/94563b8b/attachment.html -------------- next part -------------- A non-text attachment was scrubbed... Name: GClogo.png Type: image/png Size: 9393 bytes Desc: not available Url : http://www.flux.org/pipermail/linux/attachments/20100910/94563b8b/attachment.png From wordz2u at gmail.com Sat Sep 11 09:12:00 2010 From: wordz2u at gmail.com (Steve G.) Date: Sat, 11 Sep 2010 07:12:00 -0600 Subject: [Linux] Virtualization question Message-ID: I am looking to buy a discount laptop, and seems that new models, even cheap ones, have two things in common. Multiple cores and Windows 7 64x, which is fine with me as I need a windows environment for a couple of small things. However, my normal environment is Linux, currently Ubuntu, so I am thinking of running VMWare workstation or VirtualBox with Linux in it for the 99% of use. Here is the issue I am worried about - if the machine does not support Linux fully (i.e. video or some component does not work) as a host, would it be possible to use that hardware as a guest? In other words, if my graphic or camera or sound card does not work with linux directly, would I be able to use the machine as a guest. I know with full virtualization (qemu? bochs?) it is doable, but I don't want the overhead of an emulator. THX, Z. -- Check out my web site - www.words2u.net -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100911/e410e182/attachment.html From wordz2u at gmail.com Sat Sep 11 09:12:00 2010 From: wordz2u at gmail.com (Steve G.) Date: Sat, 11 Sep 2010 07:12:00 -0600 Subject: [Linux] Virtualization question Message-ID: I am looking to buy a discount laptop, and seems that new models, even cheap ones, have two things in common. Multiple cores and Windows 7 64x, which is fine with me as I need a windows environment for a couple of small things. However, my normal environment is Linux, currently Ubuntu, so I am thinking of running VMWare workstation or VirtualBox with Linux in it for the 99% of use. Here is the issue I am worried about - if the machine does not support Linux fully (i.e. video or some component does not work) as a host, would it be possible to use that hardware as a guest? In other words, if my graphic or camera or sound card does not work with linux directly, would I be able to use the machine as a guest. I know with full virtualization (qemu? bochs?) it is doable, but I don't want the overhead of an emulator. THX, Z. -- Check out my web site - www.words2u.net -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100911/e410e182/attachment-0001.html From btabor at protexx.com Sat Sep 11 09:41:05 2010 From: btabor at protexx.com (Bill Tabor) Date: Sat, 11 Sep 2010 13:41:05 +0000 Subject: [Linux] Virtualization question In-Reply-To: References: Message-ID: <2106807794-1284212466-cardhu_decombobulator_blackberry.rim.net-295702210-@bda229.bisx.prod.on.blackberry> Look into Citrix XEN. Use its kernel to run your multiple OS. Sent via BlackBerry from T-Mobile -----Original Message----- From: "Steve G." Sender: linux-bounces at flux.org Date: Sat, 11 Sep 2010 07:12:00 To: linux Reply-To: linux at flux.org Subject: [Linux] Virtualization question _______________________________________________ Linux mailing list Linux at flux.org http://www.flux.org/mailman/listinfo/linux From lou at lous-stuff.com Sat Sep 11 10:02:09 2010 From: lou at lous-stuff.com (lou) Date: Sat, 11 Sep 2010 10:02:09 -0400 Subject: [Linux] [flux] Virtualization question In-Reply-To: References: Message-ID: <4C8B8BE1.5080208@lous-stuff.com> On 09/11/2010 09:12 AM, Steve G. wrote: > Ubuntu, so I am thinking of running VMWare workstation or VirtualBox > with Linux in it for the 99% of use. I have a several year old Thinkpad and it runs Ubuntu fine with the exception of video support. Attaching external monitors is/was a pain. I have run VirtualBox (Windows version) and it worked fine. I am now back to running Windows 7 on it with VMWare workstation. To run 64 bit guests, requires the Virtualization support in hardware (VT for Intel). Mine does. You need a lot of RAM, I have 6GB and can run 3 VM's without problems. I initially installed Xen (openSUSE), but had problems with my Nvidia video chipset. Unless they have "fixed" it, Xen would not run dual monitors. Perhaps I should rephrase that, I could not make it run dual monitors. :) I finally threw in the towel and put W7 Ultimate with bitlocker on it and run Ubuntu in a VM. Works fine. Check all hardware against the support lists. Lou -- " A people that values its privileges above its principles soon loses both." - Dwight D. Eisenhower --- __@ -- _-\<,_ --- (_)/ (_) From kwan at digitalhermit.com Sat Sep 11 10:15:57 2010 From: kwan at digitalhermit.com (Kwan Lowe) Date: Sat, 11 Sep 2010 10:15:57 -0400 Subject: [Linux] Virtualization question In-Reply-To: References: Message-ID: On Sat, Sep 11, 2010 at 9:12 AM, Steve G. wrote: > I am looking to buy a discount laptop, and seems that new models, even cheap > ones, have two things in common. Multiple cores and Windows 7 64x, which is > fine with me as I need a windows environment for a couple of small things. > However, my normal environment is Linux, currently Ubuntu, so I am thinking > of running VMWare workstation or VirtualBox with Linux in it for the 99% of > use. > Here is the issue I am worried about - if the machine does not support Linux > fully (i.e. video or some component does not work) as a host, would it be > possible to use that hardware as a guest? In other words, if my graphic or > camera or sound card does not work with linux directly, would I be able to > use the machine as a guest. > I know with full virtualization (qemu? bochs?) it is doable, but I don't > want the overhead of an emulator. VMWare virtualizes the hardware, so even if Linux does not fully support a hardware feature, in the virutalization environment it will very likely be supported. For example, I had one piece of video hardware that I could never get working quite right with Linux. There were drivers for WinXP however. After loading Linux under that XP-based VMWare host, the video performance was actually better using the VMWare console than running Linux natively. That was quite a while back though. Today I am more apt to install Linux directly on the hardware and put Win7 or WinXP in a virtual machine to run the occasional Windows applications (Microsoft Office is the primary culprit). From wordz2u at gmail.com Sat Sep 11 11:14:20 2010 From: wordz2u at gmail.com (Steve G.) Date: Sat, 11 Sep 2010 09:14:20 -0600 Subject: [Linux] Virtualization question In-Reply-To: References: Message-ID: Thanks, I think these answer my question. I need Windows for a Magellan GPS that is not supported anywhere else, but the main reason to buy with Windows 7 is that the majority of discounted recent laptops are not Linux compatible or at least not Linux tested. I am scanning the weekly special offers from BestBuy, Amazon, TigerDirect, may even go Target or Walmart if I get desperate... Last week I had a chance to buy an i3 Acer for $377, but before I made up my mind supply ran out (had my reasons to wait at the time, won't happen again). Right now, the low end (under $400) are Celerons and the multi-core are $500+, but surely something worthwhile will come up soon. Will let you know how it works out when I get the machine... Z. On Sat, Sep 11, 2010 at 8:15 AM, Kwan Lowe wrote: > On Sat, Sep 11, 2010 at 9:12 AM, Steve G. wrote: > > I am looking to buy a discount laptop, and seems that new models, even > cheap > > ones, have two things in common. Multiple cores and Windows 7 64x, which > is > > fine with me as I need a windows environment for a couple of small > things. > > However, my normal environment is Linux, currently Ubuntu, so I am > thinking > > of running VMWare workstation or VirtualBox with Linux in it for the 99% > of > > use. > > Here is the issue I am worried about - if the machine does not support > Linux > > fully (i.e. video or some component does not work) as a host, would it be > > possible to use that hardware as a guest? In other words, if my graphic > or > > camera or sound card does not work with linux directly, would I be able > to > > use the machine as a guest. > > I know with full virtualization (qemu? bochs?) it is doable, but I don't > > want the overhead of an emulator. > > VMWare virtualizes the hardware, so even if Linux does not fully > support a hardware feature, in the virutalization environment it will > very likely be supported. For example, I had one piece of video > hardware that I could never get working quite right with Linux. There > were drivers for WinXP however. After loading Linux under that > XP-based VMWare host, the video performance was actually better using > the VMWare console than running Linux natively. That was quite a > while back though. > > Today I am more apt to install Linux directly on the hardware and put > Win7 or WinXP in a virtual machine to run the occasional Windows > applications (Microsoft Office is the primary culprit). > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux > -- Check out my web site - www.words2u.net -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100911/daaafb33/attachment.html From granchijp at gmail.com Sat Sep 11 16:15:21 2010 From: granchijp at gmail.com (Jean Philippe Granchi) Date: Sat, 11 Sep 2010 16:15:21 -0400 Subject: [Linux] Virtualization question In-Reply-To: References: Message-ID: <4C8BE359.3080202@gmail.com> On 09/11/2010 11:14 AM, Steve G. wrote: > that the majority of discounted recent laptops are not > Linux compatible or at least not Linux tested. > I am scanning the weekly special offers from BestBuy, Amazon, > TigerDirect, may even go Target or Walmart ... > Will let you know how it works out when I get the machine... Speaking of do you know Zareason ? They are manufacturer of Linux Desktops/Laptops. http://zareason.com/shop/home.php One way to make sure to know what is inside your computer, and make sure it is Linux compatible. They are based in Berkley CA. Also for Magellan, I had some success in the past with GpsDrive http://www.gpsdrive.de/index.shtml my 2 cents, - Jean-Philippe From dwilliams at dtw-consulting.com Thu Sep 16 11:36:34 2010 From: dwilliams at dtw-consulting.com (David Williams) Date: Thu, 16 Sep 2010 11:36:34 -0400 Subject: [Linux] Failed Hard Drive (moving /home) In-Reply-To: References: Message-ID: <4C923982.5080507@dtw-consulting.com> Ok, So, I got my replacement drive and I also bought a new drive, so I have been moving data around my different drives. Now, I really would like to move my home mount to another drive. What's the best way to do this. For the other mount points I've been using the Mandrake Control Centre to umount partitions, and then Dolphin to move files around. If I log as my user then I'm pretty sure I won't be able to umount home as it will be in use. If possible I'd rather do this through the GUI, but am ok doing this trough the command line if needed, but might need some help. Thanks. ------------------------------------------------------------------------ David Williams Check out our WebOS mobile phone app for the Palm Pre and Pixi: Golf Caddie | Golf Caddie Forum | Golf Caddie FAQ by DTW-Consulting, Inc. On 9/9/2010 9:53 AM, CarlC - Information wrote: > > >> -----Original Message----- >> Date: Thu, 09 Sep 2010 08:50:36 -0400 >> From: David Williams >> Yes, I tried to run fsck on /dev/sdc and /dev/sdc1. I put >> that in my original post. Didn't work. >> Might try the freezer thing then....worth a shot I guess. > The real trick is to first use smartd/smartctl. Using smartctl, you can see if the disk is truly gone: > > Smartctl -a /dev/sdc | more > > Look at the ECC and Raw_Read_Error_Rate. Also, the Seek_Error_Rate will show if your taking seek errors as well. > > If Adam ever wants, I can give an entire presentation on Disk drive monitoring and health. Smartd/smartctl is your saviour :) . > Especially when you can use smartd to monitor your drives, test them when you want, and even test individual drives in a Raid set > (provided the Raid controller allows it). > > I have yet to lose a drive without smartd warning me, usually weeks to months in advanced. For my Hosting business, it's saved me > so much time as it's one of the best ways to be in preventative maintenance mode. > > As to recovery, between and ice pack/heating pad, and luck, well, now we know why backups are critical. Professional shops can > take the drive appart and use tricks like off-center head readings to get the data back, but then it's all a matter of how much > money is the data worth. > > Carl > > P.s. if your drive is USB, smartd may not work. You may have to take the drive out of the USB containter and directly connect it. > I've even seen a case where the USB controller died, but the drive was fine and we could easily recover the data while the drive > was in direct connection. > > _______________________________________________ > Linux mailing list > Linux at flux.org > http://www.flux.org/mailman/listinfo/linux -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100916/563abb71/attachment.html -------------- next part -------------- A non-text attachment was scrubbed... Name: GClogo.png Type: image/png Size: 9393 bytes Desc: not available Url : http://www.flux.org/pipermail/linux/attachments/20100916/563abb71/attachment.png From info at carlc.com Fri Sep 17 09:23:08 2010 From: info at carlc.com (CarlC - Information) Date: Fri, 17 Sep 2010 09:23:08 -0400 Subject: [Linux] Failed Hard Drive (moving /home) In-Reply-To: References: Message-ID: > -----Original Message----- > Date: Thu, 16 Sep 2010 11:36:34 -0400 > From: David Williams > > Ok, > > So, I got my replacement drive and I also bought a new drive, > so I have been moving data around my different drives. > Now, I really would like to move my home mount to another > drive. What's the best way to do this. For the other mount > points I've been using the Mandrake Control Centre to umount > partitions, and then Dolphin to move files around. If I log > as my user then I'm pretty sure I won't be able to umount > home as it will be in use. > > If possible I'd rather do this through the GUI, but am ok > doing this trough the command line if needed, but might need > some help. > > Thanks. 1) Connect new HD with PC off. Turn on the PC, make sure you can see the new /dev/sdX device. 2) Create the new partition on the /dev/sdX device to be home. This could be something like: create parition, example: parted or fdisk mke2fs -L home -j -t ext4 /dev/sdXY (where X=new drive letter and Y=Partition number) 3) Make your changes in /etc/fstab for the new device. 4) Boot your system using a separate CD 5) Mount your old drives home parition (assume it's /dev/sda7 for fun) mkdir /mnt/old mount /dev/sda7 /mnt/old 6) Mount your new drive home partition (assume it's /dev/sdb1 for fun) mkdir /mnt/new mount /dev/sdb1 /mnt/new 7) Copy over everything cp -ax /mnt/old/* /mnt/new 8) Unmount everything (same as doing a 'sync' a few times, ensures everything is written) 9) boot system NOTE: since this was done in two minutes, I may have missed a few steps or made a mistake or two. So please, tell us where your stuck, and we can help you get around it. Also, you can look at: http://www.faqs.org/docs/Linux-mini/Hard-Disk-Upgrade.html While it's old, the general principal is the same $0.02, Carl From flux.simicich at spamgourmet.com Fri Sep 17 12:55:30 2010 From: flux.simicich at spamgourmet.com (flux.simicich at spamgourmet.com) Date: Fri, 17 Sep 2010 12:55:30 -0400 Subject: [Linux] HP 2133 Mini note - lost disk password Message-ID: Someone I know bought an HP 2133 mini note that they proceeded to set a hard drive password on. The keyboard failed and they could not use the device for some time. I repaired the keyboard, (dirty zif connector) but they have forgotten the disk password through disuse. As far as I can tell, the only way to recover from this situation is to replace the hard drive. They are willing to do that. This becomes interesting, as I simply can't figure out how to write suse on a new hard drive I'l like to write the official HP release with an image of the official HP recovery partition. I'd settle for a bootable suse that was patched with the HP mods. Usually there is a recovery disk that HP creates, but I don't see one anywhere on the HP web site. HP includes the srpms for what they have changed, but that seems to be all. I think that I should install the new hard drive into a usb enclosure and stick it on another linux box, and then try and prep it from there using whatever I can find of Suse 10 and the srpms. But the system seems to have only two booting options, pxe - hard wired network, and internal hard disk. No options to boot from USB appear in the list, and no CD is installed. Anyone have any advice about this matter? I'd like to have a plan before I suggest that they buy a hard drive. If anyone knows that HP has a recovery disk for suse that I could buy from them, I'd do that. I've thoroughly searched their web site and found nothing. -- Life is too short to eat an egg white omelette. Fucking oatmeal tastes better. Order that. Only jackasses order egg white omelettes. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.flux.org/pipermail/linux/attachments/20100917/2507d9ad/attachment.html