Jump to content

iMessage not working - iCloud/Internet etc working - Changed boot.plist file and network.plist file - Help


  • Please log in to reply
418 replies to this topic

#161
fffeee

fffeee

    InsanelyMac Protégé

  • Members
  • PipPip
  • 64 posts

good-samaritan-message
That sounds interesting.

rcork can u do that w/ refit please?


good-samaritan-message is the greeting you can put in the login screen. e.g. mine says "$name | +phonenumber reward if found"

it needs to be in nvram so that the EFI boot environment can display it before filevault is unlocked.

VMware would fix the bug via a software update to their product rather than instruct on how to fix it inside the system, meaning you'd only get it working through VMs (as would be the only thing they care about). Meaning you'd have a lot of work on your hands regardless of their fix, if only to figure out how they fixed it in the first place.

Right?


That depends entirely on what method they use to resolve the issue. If they have to pass something from the host to the guest, yes, but if they have to expose the method used to generate the values for a Guest or via VMWare Tools, no.
  • p.H likes this

#162
cosmo1t

cosmo1t

    InsanelyMac Protégé

  • Developers
  • 98 posts
  • Gender:Male
i'm not entirely sure this an nvram issue.. i think it's going to take some serious dbugging (like replacing functions that are being called to dump the data).
will keep trying.

#163
flux84

flux84

    InsanelyMac Protégé

  • Members
  • Pip
  • 8 posts
I was able to sign in and register this morning after tinkering with it quite a bit. The thread over on ######## has the details on how to get it working.

#164
cosmo1t

cosmo1t

    InsanelyMac Protégé

  • Developers
  • 98 posts
  • Gender:Male
this is from other site and tflux

Ok! Huge progress to report. I was able to sign into iMessage and register!
Attached File  Screen Shot 2013-01-08 at 11.19.43 AM.png   70.99KB   51 downloads

Here is what I learned:
  • My serial number is set to a real MacPro 3,1 serial (which is what I'm set up as in SMBIOS)
  • I manually set the NVRAM MLB variable to the board serial of a real MacPro3,1
  • I manually set the NVRAM ROM variable to my MAC address (Gigabyte onboard NIC, not Apple)
  • My IOPower values for Gq3489ugfi, Fyp98tpgj and kbjfrfpoJU are all set, but do not correspond to what a real MacPro report
  • My MAC address is that of a Gigabyte NIC, not Apple
  • My Hardware UUID is as it always was, and does not correspond to a real MAC
  • The problem was my IOPower values in NVRAM where blank - I tested using real values from a MacPro3,1 by the method described below:
  • oycqAZloTNDm: <00000000 00000000 00000000 00000000 00>
  • abKPld1EcMni: <00000000 00000000 00000000 00000000 00>
I did a quick little hack in GDB to set these two values and was able to sign in without an issue. Here is how I did that:
  • Set those 2 values in NVRAM using the nvram command (effectively placing them in IODeviceTree:/options)
  • I put a breakpoints at the instructions 0x39c18 and 0x39c8c from IMDAppleServices (disassemble in Hopper demo to see what these are)
  • Began to sign into iMessage, breakpoint #1 eventually trips
  • Looked at the stack to see what was being placed into %rsi (the value pointing to IODeviceTree:/options) and continued
  • Breakpoint #2 trips moments later, but I've stopped after it has placed the IOPower pointer into %rsi, at which point I replace it with what I found in the previous breakpoint. Effectively making IMDAppleServices pull the values from IODeviceTree:/options (where I can easily set it), as opposed to IOPower.
  • Viola - iMessage signs in!
Long story short:
  • IOPower Gq3489ugfi, Fyp98tpgj and kbjfrfpoJU are all used and require values, but the value doesn't seem to matter - whatever is setting them now seems to be doing so appropriately.
  • SMBIOS values I have set include: SMfamily, SMproductname, SMboardproduct, SMserial, SMbiosversion, SMmanufacturer, SMbiosvendor. Based on verbose logging of imagent, the family, product name, board product, and serial are used for sure.
  • Hardware UUID is used, but value doesn't seem to matter
  • ROM and MLB NVRAM values are required and seem to trigger the new authentication method as opposed to the Beta method. The ROM value seems to be happy if it matches your primary network interface. Not sure about MLB - it is definitely the board serial, but whether it must be valid, or further to that - match your platform serial in some way, I'm not sure.
  • IOPower NVRAM values oycqAZloTNDm and abKPld1EcMni must be set - no idea what they are however.


#165
rcork

rcork

    InsanelyMac Geek

  • Members
  • PipPipPip
  • 167 posts
Very interesting. Looks like we need to start compiling values for the IOPower variables from different mac models or just use the MacPro3,1 values.

#166
flux84

flux84

    InsanelyMac Protégé

  • Members
  • Pip
  • 8 posts
Not necessarily... only 2 of my 5 IOPower variables matched a real MacPro. The first 3 were set at boot time and evidently worked fine. Figuring out what triggers the last 2 to be set (by the kernel maybe?) may be the best option.

#167
fffeee

fffeee

    InsanelyMac Protégé

  • Members
  • PipPip
  • 64 posts
It seems likely that these are uniquely identifying but I don't know how easily they can be reversed or if they are a one-way ticket.

Are people going to be willing to share them?

or better yet, are they maybe not even validated? (i.e. could they be anything that matches the format?)

#168
flux84

flux84

    InsanelyMac Protégé

  • Members
  • Pip
  • 8 posts

or better yet, are they maybe not even validated? (i.e. could they be anything that matches the format?)


I think this is a logical next step, but having some easy method to set these variables to test them would be incredibly helpful. Overwriting registers somewhat blindly in GDB to trick IMDAppleServices to pull the values from somewhere else is a bit tedious.

#169
rcork

rcork

    InsanelyMac Geek

  • Members
  • PipPipPip
  • 167 posts

I think this is a logical next step, but having some easy method to set these variables to test them would be incredibly helpful. Overwriting registers somewhat blindly in GDB to trick IMDAppleServices to pull the values from somewhere else is a bit tedious.


How can we inspect these variables on real macs?

#170
flux84

flux84

    InsanelyMac Protégé

  • Members
  • Pip
  • 8 posts
The code attached (courtesy of ElNono and mdmwii from ######## forum, plus 2 minor corrections from me) will do it. Compile it in XCode as a Foundation console app, needs to link against IOKit.Framework as well.

Should get output like this:

2013-01-07 16:26:18.736 checkMessages[7937:707] Gq3489ugfi: <value here>
2013-01-07 16:26:18.738 checkMessages[7937:707] Fyp98tpgj: <value here>
2013-01-07 16:26:18.739 checkMessages[7937:707] kbjfrfpoJU: <value here>
2013-01-07 16:26:18.739 checkMessages[7937:707] IOPlatformSerialNumber: value here
2013-01-07 16:26:18.740 checkMessages[7937:707] IOPlatformUUID: value here
2013-01-07 16:26:18.740 checkMessages[7937:707] board-id: Mac-F42C88C8
2013-01-07 16:26:18.741 checkMessages[7937:707] product-name: MacPro3,1
2013-01-07 16:26:18.741 checkMessages[7937:707] 4D1EDE05-38C7-4A6A-9CC6-4BCCA8B38C14:ROM: <value here>
2013-01-07 16:26:18.742 checkMessages[7937:707] 4D1EDE05-38C7-4A6A-9CC6-4BCCA8B38C14:MLB: value here
2013-01-07 16:26:18.742 checkMessages[7937:707] oycqAZloTNDm: <value here>
2013-01-07 16:26:18.743 checkMessages[7937:707] abKPld1EcMni: <value here>

Attached Files



#171
fffeee

fffeee

    InsanelyMac Protégé

  • Members
  • PipPip
  • 64 posts
I have values for oycqAZloTNDm and abKPld1EcMni that persist across reboots on my MacMini5,3 and MacBookPro10,1, I just don't know the sanest way to share them. Any suggestions? I'm inclined to just leaving them unattributed without the key values as a gist. It looks like the thread on uggc://jjj.gbalznpk86.pbz/ is looking at it from the other end, rather than attempting to reverse the format they're thinking of just finding what happens when it is set?

This may also provide some clues as to why they have implemented it this way.

#172
rcork

rcork

    InsanelyMac Geek

  • Members
  • PipPipPip
  • 167 posts
Anyone know whether Clover lets you configure these IOPower variables?

#173
p.H

p.H

    InsanelyMac Legend

  • FAQ Team
  • 740 posts
  • Gender:Male
  • Interests:Hackintosh & NBA & COD4 promod

I don't think so because it doesn't work in vmware

Message in VM still works fine?

#174
Pene

Pene

    InsanelyMac Geek

  • Coders
  • 104 posts
  • Gender:Male

Message in VM still works fine?

No it does not. Stopped working when it stopped for hacks.

#175
Slice

Slice

    InsanelyMacaholic

  • Local Moderators
  • 2,896 posts
  • Gender:Male
  • Location:Moscow

Anyone know whether Clover lets you configure these IOPower variables?

I can do this if you explain me more carefully what to do.

#176
dannymichel

dannymichel

    InsanelyMac Sage

  • Members
  • PipPipPipPipPip
  • 298 posts
  • Gender:Male

Here's my solution: Not perfect and won't work for everyone. You need a jailbroken iOS device. Download and run "Remote Messages" from cydia. I only have iPad it it's iPhone only, had to download from Mac and transfer it via SFTP then use iFile to install. Once installed and enabled in settings, you can use any web browser on your local network to use iMessage. This also lets me use it on my Lion only MacBook and could use it on Windows if I used Windows. Not a fix, but a good enough work around for me at the moment assuming it all works like it should.

Argh, even though this is only a workaround, I'm cursing the jailbreak gods. I lost my jailbreak when upgrading to iOS 6

#177
clubber_77

clubber_77

    InsanelyMac Protégé

  • Members
  • Pip
  • 33 posts
iMessage works fine with Clover rev979.
Add to config plist

<key>RtVariables</key>
<dict>
<key>MLB</key>
<string>UseYourSerialNumber</string>
<key>ROM</key>
<data>
MacAddressEthernetCard
</data>
</dict>


#178
sekmo

sekmo

    InsanelyMac Protégé

  • Members
  • Pip
  • 18 posts
Would you say that I just need my macbook pro serial number and mac address?? :weight_lift: :weight_lift:

#179
clubber_77

clubber_77

    InsanelyMac Protégé

  • Members
  • Pip
  • 33 posts

Would you say that I just need my macbook pro serial number and mac address?? :weight_lift: :weight_lift:

yes

#180
dannymichel

dannymichel

    InsanelyMac Sage

  • Members
  • PipPipPipPipPip
  • 298 posts
  • Gender:Male

yes

So that's the fix? We all need Macbook Pros to fix out hackintosh computers designed to not need Macs?





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users

© 2014 InsanelyMac  |   News  |   Forum  |   Downloads  |   OSx86 Wiki  |   Mac Netbook  |   PHP hosting by CatN  |   Designed by Ed Gain  |   Logo by irfan  |   Privacy Policy