User Tag List

  1. Hogs4Life's Avatar
    don`t fight in here, let`s stay on the subject pls.
    Tryin'.
    2010-05-13 07:26 AM
  2. Copyrighted's Avatar
    Anyway i got stuck in Apple Logo after i did a network reset since then everything worked fine.
    2010-05-13 07:42 AM
  3. iBeej's Avatar
    Hey all, sorry for the delay. Here is today's news:

    Some of you will become so entirely irritated with me when I tell you I intentionally broke my own iPhone by UNINSTALLING diskdev-cmds and then manually going through /sbin and arbitrarily deleting binaries. I can see how this would be frustrating, because here you guys are, desperately wanting out of that mess, and I go and do it on purpose. There is good reason for it though, because I get instant feedback on test results and will ensure I have done everything I can to make sure it works properly, safely, and reliably. Plus, I don't have a working phone either, so it's motivating.

    I successfully patched in the payload through the disassembled code and rebuilt the exe (manual compiling lol) and came up with 3 windows versions piggybacking off blackra1n. I have 4 testers working with me and their results provided some valuable clues in to why the windows version didn't succeed. The first problem was my attempt at slimming the payload down to the core binaries for /sbin that I assumed would be the only ones MISSING from your sbin. After seeing SOME results from the windows patch, I started to get a confirmation this is far worse than I had originally thought.

    It appears the RockApp Update (which i'll call it from now on) caused one hell of a mess. I was able to confirm my fears when I got my hands on the diskdev-cmds deb directly from the cydia repository. I manually unpacked the contents to inspect the contents and determine just how many binaries could be broken. It turns out, there are a LOT. And not just in /sbin, but also /usr/sbin. I can't say for SURE that the /usr/sbin packages are missing from your phones now, but I can safely assume this is most likely the case if the update failed and "wiped" each binary it was SUPPOSED to replace. This explains a couple things:

    1.) For the users who successfully patched sbin with my redsn0w release, something in their configuration spared them the failure in /usr/sbin.

    2.) For the other users who patched sbin, and noticed it effected how the phone was booting (getting further) but failing again, my theory is their /usr/sbin is broken too. AND I think missing a few others inside /sbin.

    I don't believe it will be possible to determine why this is different between users, but it is what it is. So I went back and rebuilt the payload with EVERYTHING from diskdev-cmds PLUS what was supposed to be in /sbin already. This is precisely why the windows version wasn't released today, because the payload went from 262k, to 1.4MEG. AND THAT'S COMPRESSED!! That gives you an idea just how severe this is. Right about now, your blood is probably boiling realizing just how bad that update blew your iphone's brains all over the wall.

    Geohot built a fixed 512kb purpled1sk.DMG image which is embedded in the executable. So again, because the payload I need to patch in to your phones is larger than that, it requires an exceptional amount of disassembly and mind grueling work to forcefully expand the EXE and update every single .data, .rdata, .text, .code, .bss., .idata, and .rsrc code section of the EXE. For a brief explanation for what this means... those "sections" are blocks of compiled code in the EXE and heavily depend on the EXE being the fixed size it is because each and every one of them also has a "fixed" register (memory address) to themselves and each other. When I go in there and stuff in a larger payload which increases the size of the EXE, all of these sections are now pointing to completely arbitray locations, which effectively "corrupts" the EXE. I have to painstakingly iron all of this out manually one byte at a time while staring at hexadecimal for hours. THIS is what slowed the windows release down today.

    However, I am DEBATING whether or not I simply split the payload up between two different EXE's. You would have to run the first one, then the second one. But doing this would allow me to push this out faster, even though it's not as slick as a one-time run. LOL, but with how fast blackra1n is, you would still be done faster than the Mac Redsn0w version.

    SO, we are close. Very close. The windows version isn't a concept or dream. I HAVE working executables. And 4 of you have already seen it. Now it's just working out some logistics and fighting my way through code, YET AGAIN. So hang in there... I haven't given up. In fact, i'm spending more time on this than you can imagine. Any minute, any hour, maybe any day now... we should be fixing phones left and right.

    Stay tuned...


    BEEJ
    Last edited by iBeej; 2010-05-28 at 08:45 PM.
    2010-05-13 07:57 AM
  4. z1mp's Avatar
    I really don't care.

    You can spin it to sound however you want, its a message board, not hard to do.

    Now all you're doing is crying and being a pain in my side, but luckily ignore fixes that ya know?

    Again, I could really care less what you think/say so save your time typing out your spin stories to me mkay?
    Get trolled.

    Hey all, sorry for the delay. Here is today's news:

    Some of you will become so entirely irritated with me when I tell you I intentionally broke my own iPhone by UNINSTALLING diskdev-cmds and then manually going through /sbin and arbitrarily deleting binaries. I can see how this would be frustrating, because here you guys are, desperately wanting out of that mess, and I go and do it on purpose. There is good reason for it though, because I get instant feedback on test results and will ensure I have done everything I can to make sure it works properly, safely, and reliably. Plus, I don't have a working phone either, so it's motivating.

    I successfully patched in the payload through the disassembled code and rebuilt the exe (manual compiling lol) and came up with 3 windows versions piggybacking off blackra1n. I have 4 testers working with me and their results provided some valuable clues in to why the windows version didn't succeed. The first problem was my attempt at slimming the payload down to the core binaries for /sbin that I assumed would be the only ones MISSING from your sbin. After seeing SOME results from the windows patch, I started to get a confirmation this is far worse than I had originally thought.

    It appears the RockApp Update (which i'll call it from now on) caused one hell of a mess. I was able to confirm my fears when I got my hands on the diskdev-cmds deb directly from the cydia repository. I manually unpacked the contents to inspect the contents and determine just how many binaries could be broken. It turns out, there are a LOT. And not just in /sbin, but also /usr/sbin. I can't say for SURE that the /usr/sbin packages are missing from your phones now, but I can safely assume this is most likely the case if the update failed and "wiped" each binary it was SUPPOSED to replace. This explains a couple things:

    1.) For the users who successfully patched sbin with my redsn0w release, something in their configuration spared them the failure in /usr/sbin.

    2.) For the other users who patched sbin, and noticed it effected how the phone was booting (getting further) but failing again, my theory is their /usr/sbin is broken too. AND I think missing a few others inside /sbin.

    I don't believe it will be possible to determine why this is different between users, but it is what it is. So I went back and rebuilt the payload with EVERYTHING from diskdev-cmds PLUS what was supposed to be in /sbin already. This is precisely why the windows version wasn't released today, because the payload went from 262k, to 1.4MEG. AND THAT'S COMPRESSED!! That gives you an idea just how severe this is. Right about now, your blood is probably boiling realizing just how bad that update blew your iphone's brains all over the wall.

    Geohot built a fixed 512kb purpled1sk.DMG image which is embedded in the executable. So again, because the payload I need to patch in to your phones is larger than that, it requires an exceptional amount of disassembly and mind grueling work to forcefully expand the EXE and update every single .data, .rdata, .text, .code, .bss., .idata, and .rsrc code section of the EXE. For a brief explanation for what this means... those "sections" are blocks of compiled code in the EXE and heavily depend on the EXE being the fixed size it is because each and every one of them also has a "fixed" register (memory address) to themselves and each other. When I go in there and stuff in a larger payload which increases the size of the EXE, all of these sections are now pointing to completely arbitray locations, which effectively "corrupts" the EXE. I have to painstakingly iron all of this out manually one byte at a time while staring at hexadecimal for hours. THIS is what slowed the windows release down today.

    However, I am DEBATING whether or not I simply split the payload up between two different EXE's. You would have to run the first one, then the second one. But doing this would allow me to push this out faster, even though it's not as slick as a one-time run. LOL, but with how fast blackra1n is, you would still be done faster than the Mac Redsn0w version.

    SO, we are close. Very close. The windows version isn't a concept or dream. I HAVE working executables. And 4 of you have already seen it. Now it's just working out some logistics and fighting my way through code, YET AGAIN. So hang in there... I haven't given up. In fact, i'm spending more time on this than you can imagine. Any minute, any hour, maybe any day now... we should be fixing phones left and right.

    Stay tuned...
    Awesome man keep it up most of us are waiting patiently.
    Last edited by z1mp; 2010-05-13 at 08:03 AM. Reason: Automerged Doublepost
    2010-05-13 08:03 AM
  5. iBeej's Avatar
    Hey iBeej... First of all, despite my lack of success, I do want to say I greatly appreciate your efforts on this...

    I tried your solution and unfortunately it didn't work for me... Here's my details:

    iPhone firmware 3.1.2
    jailbreak was blackra1n
    I use windows, but used a Mac to apply your fix in case that's relevant...


    Before your fix, my iPhone showed the apple logo for 10-15 seconds, then would just shut off.

    After the fix, I see the apple logo for maybe 10 seconds, then the screen flashes white, and it reboots - this will keep happening forever unless I force the phone to power off by holding sleep/home.

    Because it's only alive for a few seconds, I couldn't try your SSH fix suggestion - there's no way it's booted up enough to get an IP address...

    Any assistance you can provide to those of us experiencing this would be GREATLY appreciated.
    I believe I know what is happening here, even with the mac version. I mentioned above it appears SOME users are missing even more binaries from sbin than even the large payload I included in the mac version. However, the mac version did not include a /usr/sbin patch AT ALL, and my theory is this is why some of you trying the Mac fix are still unsuccessful. I havent updated the redsn0w package yet because I have been chugging away at the Windows version. But i'll fix it soon. I may even switch the mac version from redsn0w to blackra1n.

    This just goes to show you guys how diverse every iphone configuration is and how HARD it is trying to write software to accomodate all scenarios. It gives you a headache really...


    BEEJ
    Last edited by iBeej; 2010-05-28 at 08:45 PM.
    2010-05-13 08:07 AM
  6. Hogs4Life's Avatar
    Get trolled.



    Awesome man keep it up most of us are waiting patiently.
    Lol, says the guy with less posts and just joined.

    Smart comment, troll.
    2010-05-13 08:25 AM
  7. Copyrighted's Avatar
    Hogs4Life and z1mp u won`t solve anything if u keep "fighting"
    2010-05-13 08:30 AM
  8. iBeej's Avatar
    I HAVE UPDATED THE MAC VERSION REDSN0W!!! 05/13/2010 3:15AM

    I forgot I use my alarm in my iPhone, so I had to use my trusty redsn0w fixer to fix my phone rofl. So while I was at it, I repackaged it with additional binaries for /sbin as well as /usr/sbin.

    So for those of you MAC users who still had boot failures using redsn0w before. Please RE-DOWNLOAD the file and try again. Let's see how this works. I didn't change the file name, but here it is again if you're lazy.

    **REMOVED - DEPRACATED. Download new packages listed on my original post **

    I need to hit bed, but really really want to have this windows version wrapped up soon. Please keep me posted on your success with the updated redsn0w. I also have a solution for mobilesubstrate problems and can patch your springboard to disable jailbreak if that becomes an issue.


    BEEJ
    Last edited by iBeej; 2010-05-28 at 08:44 PM.
    2010-05-13 10:20 AM
  9. Copyrighted's Avatar
    i`ll wait the windows version to test it ... u did great work iBeej!
    2010-05-13 10:25 AM
  10. powermad's Avatar
    Deej What can we say but Thanks for all your effort, I will also have to try the windows version (unfortunately) since I have a PPC iMac G5 . Please dont hesitate if there is any help myself or im sure others here can offer you. Im not without a phone so im OK for now...

    Thanks Again, get some rest and I look forward to following your progress

    PS Donation is on its way! Unfortunately I cant donate SLEEP!
    How about some Ambien instead?
    2010-05-13 11:33 AM
  11. ayeayre's Avatar
    hi, thanks for the hard work.
    but, I tried the new mac version and ran into some problems, as soon as redsn0w detected DFU mode, it crashed, my iphone and redsn0w just shut down, retried a few times, i copied the errors in the console, i guess it may help determine what went wrong, it's not my mac and I'm not familiar with them so I can't give any more info, i found the console/errors by accident, all I know is its a mac book pro (fairly new) My iphone is a 3G running 3.0 and was having the rebooting problem after I tried the previous version of your fix, originally jailbroken with redsn0w and RockApp was installed, no unlock was performed.

    Instead of clogging up the thread with a long paste of errors I have pasted the log at PasteBay.com - Free uncensored text hosting

    hope it helps..
    2010-05-13 12:04 PM
  12. melaw's Avatar
    Hey all, sorry for the delay. Here is today's news:

    Some of you will become so entirely irritated with me when I tell you I intentionally broke my own iPhone by UNINSTALLING diskdev-cmds and then manually going through /sbin and arbitrarily deleting binaries. I can see how this would be frustrating, because here you guys are, desperately wanting out of that mess, and I go and do it on purpose. There is good reason for it though, because I get instant feedback on test results and will ensure I have done everything I can to make sure it works properly, safely, and reliably. Plus, I don't have a working phone either, so it's motivating.

    I successfully patched in the payload through the disassembled code and rebuilt the exe (manual compiling lol) and came up with 3 windows versions piggybacking off blackra1n. I have 4 testers working with me and their results provided some valuable clues in to why the windows version didn't succeed. The first problem was my attempt at slimming the payload down to the core binaries for /sbin that I assumed would be the only ones MISSING from your sbin. After seeing SOME results from the windows patch, I started to get a confirmation this is far worse than I had originally thought.

    It appears the RockApp Update (which i'll call it from now on) caused one hell of a mess. I was able to confirm my fears when I got my hands on the diskdev-cmds deb directly from the cydia repository. I manually unpacked the contents to inspect the contents and determine just how many binaries could be broken. It turns out, there are a LOT. And not just in /sbin, but also /usr/sbin. I can't say for SURE that the /usr/sbin packages are missing from your phones now, but I can safely assume this is most likely the case if the update failed and "wiped" each binary it was SUPPOSED to replace. This explains a couple things:

    1.) For the users who successfully patched sbin with my redsn0w release, something in their configuration spared them the failure in /usr/sbin.

    2.) For the other users who patched sbin, and noticed it effected how the phone was booting (getting further) but failing again, my theory is their /usr/sbin is broken too. AND I think missing a few others inside /sbin.

    I don't believe it will be possible to determine why this is different between users, but it is what it is. So I went back and rebuilt the payload with EVERYTHING from diskdev-cmds PLUS what was supposed to be in /sbin already. This is precisely why the windows version wasn't released today, because the payload went from 262k, to 1.4MEG. AND THAT'S COMPRESSED!! That gives you an idea just how severe this is. Right about now, your blood is probably boiling realizing just how bad that update blew your iphone's brains all over the wall.

    Geohot built a fixed 512kb purpled1sk.DMG image which is embedded in the executable. So again, because the payload I need to patch in to your phones is larger than that, it requires an exceptional amount of disassembly and mind grueling work to forcefully expand the EXE and update every single .data, .rdata, .text, .code, .bss., .idata, and .rsrc code section of the EXE. For a brief explanation for what this means... those "sections" are blocks of compiled code in the EXE and heavily depend on the EXE being the fixed size it is because each and every one of them also has a "fixed" register (memory address) to themselves and each other. When I go in there and stuff in a larger payload which increases the size of the EXE, all of these sections are now pointing to completely arbitray locations, which effectively "corrupts" the EXE. I have to painstakingly iron all of this out manually one byte at a time while staring at hexadecimal for hours. THIS is what slowed the windows release down today.

    However, I am DEBATING whether or not I simply split the payload up between two different EXE's. You would have to run the first one, then the second one. But doing this would allow me to push this out faster, even though it's not as slick as a one-time run. LOL, but with how fast blackra1n is, you would still be done faster than the Mac Redsn0w version.

    SO, we are close. Very close. The windows version isn't a concept or dream. I HAVE working executables. And 4 of you have already seen it. Now it's just working out some logistics and fighting my way through code, YET AGAIN. So hang in there... I haven't given up. In fact, i'm spending more time on this than you can imagine. Any minute, any hour, maybe any day now... we should be fixing phones left and right.

    Stay tuned...
    Hi iBeej

    Your Redsnow recovers my iPhone and I have manage to backup all my important documents and also manage to syn to iTune to create a backup!! Thank you so much!!!

    I will donate for sure once I'm back from my business trip, please wait for my contribution.

    Below are some pointers that might help u iBeej,

    How come I can manage to recover my iPhone successfully but not some other users?

    Before my iPhone went into dead mode, I saw that there are 7 updates from Rock App, so i went to press update all but only a few updates went through but some other updates encounters errors. So i didnt bother to update again n continue using my iPhone but find that my iPhone is getting slower n slower so i restart it and thats when my iPhone is DEAD.

    So the main point is maybe some of you bother to update all 7 updates from ROCK APP that might further damage the system files in your iPhone.

    Above is what i suspect that might be the caused between those who succeeded n those who failed.
    2010-05-13 12:36 PM
  13. rilloteddy's Avatar
    hi, thanks for the hard work.
    but, I tried the new mac version and ran into some problems, as soon as redsn0w detected DFU mode, it crashed....

    Hi again BEEJ!

    I tried your new mac-version of redsnow but as in the case with ayeayere, redsnow crashed.
    Though my iPhone is in recovery-mode and didn't even flinch. (Not DFU...)
    Redsnow crashes just after i press next on the "install-cydia checkbox" prompt.

    I too have a macbook pro, intel core2duo 2.66.
    My iphone is a 3GS, 3.1.2.
    I'm JB and unlocked with blackra1n.

    This is my data from iRecovery:

    [FTL:MSG] Apple NAND Driver (AND) RO
    [NAND] Found Chip ID 0x3295DE987A on FMI0:CE0
    [NAND] Found Chip ID 0x3295DE987A on FMI0:CE1
    [NAND] Found Chip ID 0x3295DE987A on FMI1:C8
    [NAND] Found Chip ID 0x3295DE987A on FMI1:CE9
    [FTL:MSG] FIL_Init [OK]
    [FTL:MSG] BUF_Init [OK]
    [FTL:MSG] FPart Init [OK]
    read new style signature 0x43313133 (line:375)
    [FTL:MSG] VSVFL Register [OK]
    [FTL:MSG] VFL Init [OK]
    [FTL:MSG] VFL_Open [OK]
    [FTL:MSG] YAFTL Register [OK]
    [FTL:MSG] FTL_Open [OK]
    Boot Failure Count: 2 Panic Fail Count: 0
    Delaying boot for 0 seconds. Hit enter to break into the command prompt...
    HFInitPartition: 0x4ffa1040
    Kernelcache image not valid
    Entering recovery mode, starting command prompt
    (Recovery) iPhone$

    Something that differs from other peoples (with the same prob.) iRecovery-info is the lines -

    HFInitPartition: 0x4ffa1040
    Kernelcache image not valid

    And that doesn't sound good? lol

    I'm very very thankful 4 all your effort! Your dedication rings thru the interweb...

    Thanx!
    /Richard
    2010-05-13 01:36 PM
  14. iBeej's Avatar
    That's the other problem. Kernel cache. I have some theories we can try. Hang in there. We are one step closer.

    BEEJ
    Last edited by iBeej; 2010-05-28 at 08:41 PM.
    2010-05-13 04:33 PM
  15. Copyrighted's Avatar
    Good work, if i can help u with testing, notice me. I`m online allmost everyday even if i`m home or at work.
    2010-05-13 04:55 PM
  16. icegunner's Avatar
    Ditto on Copyrighted. I'm available all day.
    2010-05-13 04:56 PM
  17. czxxx's Avatar
    Hey iBeej,

    I tried the new redsn0w and it crashes just as it starts to deploy to my phone. I've tried it on a MacBook Pro w/ 10.5.8 and an iMac w/ 10.6.3, same results.

    I want to reiterate once more how much I appreciate your efforts! I will be donating and I encourage others that you've helped to do the same.

    EDIT:
    My console gives the same output that ayeayre is getting.. FYI
    Last edited by czxxx; 2010-05-13 at 07:56 PM.
    2010-05-13 07:20 PM
  18. 4thMeal's Avatar
    Hi iBeej,

    Boy was I glad to see someone was looking for a solution rather than just restoring from backup.

    1.) iPhone OS firmware version
    3.1.3

    2.) Unlocked or not
    Not unlocked

    3.) Which jailbreak did you use
    sn0wbreeze

    4.) Do you use windows or mac.
    Both - used windows for original jailbreak

    5.) Do you at LEAST have access to a mac.
    Yes

    6.) The results of the applied FIX.
    Steps I took since I encountered the sbin issue.

    a. Found a forum post saying to update the files located in /sbin and usr/sbin. I did this without ensuring the permissions were correct and this caused my iPhone to start rebooting indefinitely.

    b. Found this forum and downloaded your modified version of redsn0w. I downloaded the newest version which included the usr/sbin files. When I ran this (using 3.1.2 firmware b/c 3.1.3 doesn't work with 0.9.3 or 0.9.4) it would get to the point where it starts the jailbreak process and then redsn0w would crash. I tried this 3 times and it happened every time. I assumed it had something to do with 0.9.3 not being able to jailbreak 3.1.3.

    c. So I read your article about modifying Cydia.tar (good article btw). I copied sbin and /usr/sbin and modified the files and folders with chown root:wheel; I verified the new owner and groups. I then downloaded redsn0w 0.9.4, unzipped it, updated the tar, and rezipped it. I ran the newly updated package and used firmware 3.1.2 (per the instructions in step 2 Jailbreak iPhone 3.1.3 with Redsn0w 0.9.4 | Redmond Pie ). Everything appeared to go smoothly. I saw the pineapple and it said synching filesystem and then it said rebooting. On reboot I noticed my sn0wbreeze icon was gone and was back to the stock apple logo however the repeating reboot continues.

    I feel like I'm close but I must be missing a step somewhere. Any pointers.

    Thanks for all of the hard work you've done!
    2010-05-13 07:33 PM
  19. spooky655's Avatar
    1.) iPod Touch 2G OS firmware version
    3.1.2

    2.) Unlocked or not
    NotUnlocked

    3.) Which jailbreak did you use
    redsn0w

    4.) Do you use windows or mac.
    Windows

    5.) Do you at LEAST have access to a mac.
    No

    I can't thank you enough for your work!!! Looking forward to a fix, you're a life saver
    2010-05-13 08:37 PM
  20. icegunner's Avatar
    I hate to sound like I'm nagging...
    But, do we have an update?
    I am still available for testing if need be.
    2010-05-13 09:56 PM
983 ... 67891018 ...
LINK TO POST COPIED TO CLIPBOARD