jaclaz

My personal preference is for dd_rescue: http://www.garloff.de/kurt/linux/ddrescue/ used TOGETHER with dd_rhelp http://www.kalysto.org/utilities/dd_rhelp/index.en.html the latter is very well documented and extremely simple to use, in practice it takes the manual intervention out from using dd_rescue directly. jaclaz

I really don't want to make a case out of it , but really, if your friend used "plain" dd on a disk that has problems, he cannot be this great Linux expert, as *any* Linux user knows (or should know) about two specific dd versions for recovery: ddrescue: http://www.gnu.org/software/ddrescue/ddrescue.html dd_rescue: http://www.garloff.de/kurt/linux/ddrescue/ jaclaz

From what you say there must be some issue with making the USB stick bootable, and/or in the copying of the files. The blinking cursoer is typical of a "wrong" setttings in the bootsector, (wrong CHS values) and this can in some case be a fault of the HP USB utility. Can you try making the stick bootable with RMPREPUSB with grub4dos, verify it boots on your machine and then try running the "special program" on it? It doesn't make sense that it doesn't boot, the thingy should work (to create the bootable stick) oon *any* machine, not necessarily on the intended targeted server. jaclaz P.S.: EDIT: out of curiosity I did a quick check. The thingy I got contains a "plain" .iso image of a Linux system. I managed to run it in QEMU allright, not by directly loading the .iso, but using a grub4dos boot floppy with this menu.lst entry (parameters got from isolinux.cfg): title stupid HP root (cd) kernel /SYSTEM/VMLINUZ media=cdrom rw root=/dev/ram0 ramdisk_size=121260 init=/bin/init loglevel=3 ide=nodma ide=noraid pnpbios=off usrramfs=1 splash=silent showopts initrd /SYSTEM/INITRD.IMG boot I presume you can do the same on USB stick using the settings in syslinux.cfg. Then it seemingly hangs on udevd, after having printed "this is not a supported server", and It takes forever to load (several minutes), but eventually you have an option to run the update for EITHER: Flash SATA drive models - GB0250C8045, GB0500C8046, GB0750C8047 Flash SATA drive models - GB0250EAFKF, GB0500EAFJH, GB0750EAFJK, GB1000EAFJL Tthe first choice seemingly runs CP010461.scexe and second CP010456.scexe (these are files that are in the .iso under COMPONENTS) BTW the documentation says: and that the HD controller has to be set as "IDE" and NOT to "SATA" or raid. I guess you need to find a friend with some Linux knowledge and understand if these files can be run from a "plainer" Linux or if the firmware can be extracted form those .scexe files. Good hints are here: http://lackof.org/matt/hacking/proliant/Debian-ProLiant-Guide.html Most probably the good guys at HP ask you to download a huge installer only to have it execute the 4 (four) batch lines in usbcreator.cmd: xcopy %1\*.* %2\ /e /k move %2\system\initrd.img %2\ move %2\system\vmlinuz %2\ move %2\usb\syslinux.cfg %2\ that should equate to: copy whole contents of CD to USB stick (in root or possibly in a folder \boot\) then move to the same (root or folder) initrd.img, vmlinuz and syslinux.cfg.

If I were you I would remove point D) as soon as you can. jaclaz

You may find this of interest: http://homepage.ntlworld.com./jonathan.deboynepollard/FGA/problem-report-standard-litany.html jaclaz

WHICH EXACT USB installer? WHAT EXACTLY are the unpacked contents of that thingy? jaclaz

NOTHING. I will translate: EITHER your friend used dd wrongly OR the whatever he dd'ed was all 00's OR an unknown error happpened OR your actual disk is all 00's. I don't really want to somehow take advantage of your misfortune till now, BUT if you quickly review this thread you will see how it goes like this: saman0suke: I need help and advice for .... jaclaz: first thing you should use tool A using it along procedure B saman0suke: OK, I used tool A with procedure C then jaclaz: well, as said you should use tool A along procedure B saman0suke: OK, I had a friend using maybe tool D along an unknown procedure, here is the result jaclaz: result is no good Your turn. jaclaz

Yes. NEITHER ! Yes. NEVER, NEVER, NEVER, for ANY reason attempt to write a firmware that is NOT specifically targeted for the device UNLESS you are told to do so by: a representative of the manufacturer someone that really knows on the SPECIFIC device and whom you can trust blindly You want the HPG6 from HP: http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?objectID=c01612429 You may want to explore the possibilities of the OFFLINE update utility: jaclaz

You are not going to get very far if the recovered data is like this. The file you posted consists of ONLY 00's. I hope that this is only somehow "a glitch in the matrix" and that the rest of the file actually contains data . You should first thing calculate the MD5 of the image file you have in your hands and compare it with the result of using this: http://www.edenprime.com/software/epAllZeroHashCalculator.htm A suitable tool to calculate the hash on the image is http://www.nirsoft.net/utils/hash_my_files.html jaclaz

I am not sure to undersatand this as well. You install a launcher (the one you like it best), then you connect it to one of the stupid Windows 8 "buttons" or whatever they are called, and you have the equivalent of a "Start menu". jaclaz

From what you write the only thing (among the "known ones") you haven't done is clearing the G-list, but I have NO idea if it could be of ANY use or if it will brick the disk for good If I get it right you are in a BSY status and not in a LBA0 one http://www.facebook.com/note.php?note_id=152456431478526 jaclaz

If I may, if DELAYEDEXPANSION is set, you don't even need the FOR /F: @ECHO OFF SETLOCAL ENABLEEXTENSIONS SETLOCAL ENABLEDELAYEDEXPANSION Set Pets=Cats Dogs Horses Set Cats=FiFi FruFru Fluffy Set Dogs=Lassie MrMuggles RinTinTin Set Horses=MrEd Trigger For %%A In (%Pets%) Do ( For %%B In (!%%A!) Do ( For %%C In (%%B) Do ( Echo %%A %%C ) ) ) jaclaz

Are you talking of DOS or of a NT system (command.com vs. CMD.EXE)? jaclaz

Just found these: http://web.archive.org/web/20100416080308/http://www.datarecoverystory.com/2009/07/typical-data-recovery-case-for-7200-9-and-7200-10/ http://web.archive.org/web/20100325045852/http://www.datarecoverystory.com/2009/06/data-recovery-from-read-only-hdd/ that seem like confiirming what the OP posted. jaclaz

Use grub4dos (or other similar bootmanager) to boot only once (the first) from USB and then boot only from HD. Add to the Windows 7 install something auto-running when install is completed that re-sets the menu.lst to boot the USB stick next time. Or set "fixed" to boot form hard disk with a 10 seconds time out and make sure to be there during first boot to select the USB stick (the thingy can be unattended as mych as you wish, but someone must be there to insert the USB thingy and switch the PC on). jaclaz

http://www.pegtop.de/start/ http://sourceforge.net/projects/asuite/ http://www.aignes.com/psmenu.htm http://www.ugmfree.it/SyMenu.aspx http://blaze-wins.sourceforge.net/index.php http://colibri.leetspeak.org/ .... jaclaz

Well, NO, it is a GOOD idea and it is perfectly possible. The posts in this thread say THREE different things: if your CD is unattended you won't be asked to press R to get to the built in Recovery Console it is possible to ADD a separate instance of Recovery Console to XP install CD by using a bootmanager the recovery console is there allright on unattended CD's, only you won't be prompted to "press R" BUT you can access it with F10 when prompted for F6 to add drivers Some reference: Fourth one is not clear, doesn't descirbe WHAT has been attempted and the results reported are different from those in the above reference thread, which contents have been verified. Recovery is NOT Repair and Repair is NOT Recovery. Yes. http://homepage.ntlworld.com./jonathan.deboynepollard/FGA/questions-with-yes-or-no-answers.html Start here: http://reboot.pro/forum/66/ jaclaz

bootrec is documented. If you use the Vista / 7 bootrec with bootrec /FixMbr it will write Vista or 7 MBR CODE. http://support.microsoft.com/kb/927392/en-us So, if you want the XP MBR code, it is NOT a good idea. If you are OK with a system that has the Windows 7 MBR code, it's OK (though non-standard). Now, guess why MBRFIX was developed? http://www.sysint.no/nedlasting/mbrfix.htm jaclaz

I like this : http://www.reddit.com/r/technology/comments/pdg4r/windows_8s_next_victim_microsoft_kills_the_start/ jaclaz

Depending on the media used and on a few other requirements that might make the difference it should be possible - by using a bootmanager such as grub4dos to have more than one WINNT.SIF, and thus choose between the UNattended (BUT with NO recovery console) and an attended (BUT WITH recovery console) or, more simply - and this is definitely possible - add an "independent" recovery console. jaclaz

Well, there is some "canonical" order in data recovery, which basically is: So you try to recover: whole disks if above not possible, whole partition/filesystem(s) if above not possible, whole file(s) if above not possible, fragment(s) of file(s) if above not possible, you have (yet another) doorstop exactly in this order . jaclaz

Well, maybe I am a genius (possibly also absolute ), but not because of that three-liner, which is very basic. Sure it does, BUT take into consideration what Yzöwl said, there are situations where this approach is NOT 100% "foolproof" or "safe". Depending on the actual value of data, I would implement a slightly more complete check, along the lines suggested. The posted example simply skips over first result: What you could do is check the first result: The above is just an example and can be definitely bettered, it will fail for files inside a subdir (but can be easily adapted to take care of this also). jaclaz

It is more likely that it is the "real" dd (originally for Linux) or maybe dd_rescue or ddrescue (two very confusingly named forks of dd aimed to data recovery still for Linux). There is no way to suggest you an app without knowing something more about what was recovered. If it was just the beginning of second partition, i.e. it's first sector is a bootsector, the best tool you can use to attempt mounting it is IMDISK (under windows), but it is possible that some data needs to be corrected, I would rather create a dummy MBR and mount it in TESTDISK. If you post first few - say 100 or 200 - sectors of the image you have I can have a look at them and ossibly give you some more targeted advice. A suitable tool (under Windows) is dsfo, part of the DSFOK toolkit, : http://members.ozemail.com.au/~nulifetv/freezip/freeware/ Unzip everything in a directory, say C:\dsfok\ Open a command prompt, navigate to it, then issue command: dsfo C:\whatever\yourimage.img 0 102400 myfirst200.bin you will get in the directory file myfirst200.bin with size 102,400 bytes that represent first 200 sectors of your image, then compress the .bin in a .zip archive and either attache it to your next post r upload it somewhere an post a link to it. jaclaz

Good. The actual version is set in SCROLL.INI BUT the only seemingly reliable data/dates are seemingly in SETUP.INF That specific driver that worked for you has in it "Mouse Driver V1.0"-> http://it.driverscollection.com/?aid=4107506016957ad05bbd7e3cb71 e-bue Movbit drivers MovBitDrive.zip and: The one here: http://radymno.org/sterowniki/Mysz_EasyTouch_5/ has also "Mouse Driver V1.0" and files have same dates, so they should be exactly the "same thing" The Pilot mouse pointertovoid initially mentioned seems like the one here : http://driverscollection.com/?H=3%20Button%20Scroll%20Mouse&By=Memorex http://driverscollection.com/?file_id=11977 Driver3202-2362.exe with "Pilot Mouse V1.00" and: The one here: http://yo.flocky.free.fr/periph/Dual%20Souris/ has "Dual Wheel Mouse V5.2 (95/98)" and: The one here: http://www.modem-drivers.com/drivers/28/28886.htm has "Dual Wheel Mouse V5.3 (95/98)" and: There is another one that you can get here: http://web.archive.org/web/200012050024/http://www.cre.com.tw/english/index.htm vivi-p.exe that has "Scroll Mouse V2.0" and: Besides the same .inf contents, the typo in Se®ialName= makes clear that the source is the same, no matter how they are re-branded, so it seems like they are all the same thingy BUT the file LoneCrusader tested successfully (e-bue Movbit drivers) is the most recent version (notwithstanding the version numbering). jaclaz

On ES2 you need to short the "read channel" to ACCESS the terminal, but if you cannot you should get a LED:000000CC FAddr:xxxxxxxx, if you get the if you get the Error 1009 DETSEC 00006008 it should mean that you already connectted to terminal, however it means "roughly" "cannot spin up", THis can be caused by problems with contacts (see above) or drive motor stitched, again see above. If drive motor is gone or bearing really stuck, you cannot do ANYTHING (without opening the drive ). Does the drive rotate/spin? See here: DO NOT EVEN THINK to swap a PCB: (unless you also know how to swap a ROM and you have the tools/knowledge, but then you wouldn't be here asking questions ) jaclaz