Jump to content

[GUIDE] Scripted Yosemite/Mavericks Install on Gigabyte Mobos


4,696 posts in this topic

Recommended Posts

"Didn't work" doesn't really give us any information to go on. You're going to have to, at least, meet us halfway by providing all details from your end. What's happening at boot time? Can you get into the bootloader? Are you booting in "verbose" mode (-v) to see the progress? Does it "hang" at boot with no visible progress? The more details you provide, the better we can zero in on the culprit.

 

The FAQs on the front page discusses many of the common issues in boot failures.

 

MAJ

 

I boot ,i press -v, and after 4-5 seconds i see the grey Apple for 1 second, than the PC restart

That's all :wallbash:

Link to comment
Share on other sites

I finally have some time to try working on this again. I have been running the DD script from within my working 10.6.4 install onto a fresh disk. But I thought maybe I would try doing it by the USB method. I've tried twice to re-create the USB install, hoever, after creation. when I reboot, my machine hangs at bios with the usb connected. Booting without the usb in (works, then at the cham boot loader, plugging in the usb stick locks the system up again.) weird, same stick I've used for other install methods from USB?

 

any ideas? I'll try booting into 10.6.4 and then plugging it in and see if it locks up the 10.6.4

Me and Sapricot discussed the exact same problem 2 pages back, here and here.

 

It seems to be a problem between the UD5 mobo and a version of Chameleon. An older version of Chameleon still works fine. So, put r115 on your USB key (option 2 in D_D's script) to do the installation and after installation overwrite the bootloader on your hard disk with the latest chameleon from D_D's script.

 

Edit: Hehe, lol @ Sapricot :P

Link to comment
Share on other sites

septicdeath: You seem to have run into the same problem I and Josvdp have already encountered. Some boot loaders just make the UD5 freeze at POST or, as you noticed, if you plug it in after cham starts.

 

Try doing the stick with DD's script and then "update" to another version of cham, for instance r426 that has been working for me. If it works, you can just keep it for the install and then do the actual install on the internal hard drive using DD's script and the newest version of cham.

 

Edit: Three minutes! You beat me by three minutes, Josvdp :P

Link to comment
Share on other sites

lemarqq,

If you are really using the -v option at boot, you wouldn't be seeing the Apple logo.

The -v option needs to be typed out at the prompt or select it via bootloader options.

 

Are you using a sleepenabler.kext? If so, which version? It may be causing the KP.

Type at the prompt -v pmVersion=0

 

MAJ

Link to comment
Share on other sites

schwamm,

I you are not using the white JMicron SATA ports, remove the JMicron36xSATA.kext. Additionally, remove the JMicronATA.kext if you are not using the JMicron PATA port.

 

I don't think your setup is using the ATY_Init.kext, as you are using the GraphicsEnabler flag and netkas' installer, so remove it. A boot cache can't be made with it in /Extra, anyway. Netkas' installer generally adds the IDs to the ATI4800Controller.kext in S/L/E, so you may not need the LegacyATI4800Controller.kext. But, I don't really know for sure in your case.

 

The JMicron stuff is likely the culprit for the random reboots, so report back if you are still experiences issues.

 

Thanks for the help. I removed ATY_init.kext and JMicron36xSATA.kext, and it's running ok right now. I only installed those kexts because I had them in my stable 10.6.2 installation. I still need the ATA kext, I don't have a sata optical drive yet, but if that's causing problems I'll just buy a new one. Thanks again and I'll let you know if I see the problems pop up again! It only happens after a few hours of use, so it's not easy to troubleshoot...

Link to comment
Share on other sites

lemarqq,

If you are really using the -v option at boot, you wouldn't be seeing the Apple logo.

The -v option needs to be typed out at the prompt or select it via bootloader options.

 

Are you using a sleepenabler.kext? If so, which version? It may be causing the KP.

Type at the prompt -v pmVersion=0

 

MAJ

 

I'm sorry, digital_dreamer, but i have no idea what sleepenabler.kext is :)

Today i tried again, with Empire EFI.

I boot with EFI, replaced the Empire EFI CD with the DVD SL, F5 - etc, i see the DVD SL, it start to install, but after about 2-3 minutes nothing happent.Monitor just flicks off and on.

The last message i can see when installing, is about "nvram" ... i think my videocard is the problem

I have Nvidia GTX 285

I tried everything possible, it doesen't work.

The only OSX version i can install is Leopard 10.5 but there are no drivers for Video / sound, etc

You think i'll have a chance to try it this way :

I install again Leopard 10.5 , and than i try to install ( with DVD SL or from USB stick ) the version SL 10.6 over 10.5

What do you think, Digital_Dreamer ? Is a good idea ? Would that work ?

I have the second HDD - 640 GB , free. If your answer is "yes" , how should i do that ?

Do i have to make 2 partitions ? 1 for Leopard, and the second for SL ?

Thanks again. Have a good day.

Link to comment
Share on other sites

I boot with EFI, replaced the Empire EFI CD with the DVD SL, F5 - etc, i see the DVD SL, it start to install, but after about 2-3 minutes nothing happent.Monitor just flicks off and on.

This is, indeed, a graphics card setup issue and sounds much different from the previous KP after 3 seconds issue.

I don't know what bootloader the Empire setup is using, but it doesn't appear to be recognizing the NV 285. A EFI string should take care of this.

 

I have the second HDD - 640 GB , free. If your answer is "yes" , how should i do that ?

Do i have to make 2 partitions ? 1 for Leopard, and the second for SL ?

Yes, to all.

I always suggest having two partitions. If you can have a working Leopard install on one, then use it to install SL on the other.

 

MAJ

Link to comment
Share on other sites

This is, indeed, a graphics card setup issue and sounds much different from the previous KP after 3 seconds issue.

I don't know what bootloader the Empire setup is using, but it doesn't appear to be recognizing the NV 285. A EFI string should take care of this.

 

 

Yes, to all.

I always suggest having two partitions. If you can have a working Leopard install on one, then use it to install SL on the other.

 

MAJ

I'll give it another try. But, please tell me how do i add this "EFI string" ?

Where do i have to add this ? On the Empire EFI boot CD ?

Please, don't mind me, Digital_Dreamer, but i really have no idea.

:) Thanks again.

PS: i'm close, i feel that :P I will use SL, very soon.

Link to comment
Share on other sites

Me and Sapricot discussed the exact same problem 2 pages back, here and here.

 

It seems to be a problem between the UD5 mobo and a version of Chameleon. An older version of Chameleon still works fine. So, put r115 on your USB key (option 2 in D_D's script) to do the installation and after installation overwrite the bootloader on your hard disk with the latest chameleon from D_D's script.

 

Edit: Hehe, lol @ Sapricot ^_^

 

 

Thanks guys, I would have never thought to try that, r115 allowed it to boot, where I was greeted with a cannot continue memory error. So I am re-making the stick and following what you said exact, copy the extra, update with r115, allow to recreate extra, and then erasing extra and restoring the original from step 15's creation.

 

Between then and now, I have been using my stable 10.6.4 (kakewalk) install to try running the script and install to another harddrive in the system. No matter what I try or who I use (tonymac's [url="http://www.insanelymac.com/forum/topic/279450-why-insanelymac-does-not-support-tonymacx86/"]#####[/url] method, DD's script with chameleon as well as pci-efi 10.5/6, Kakewalk 3.1) they all do the same thing, although most of them actually KP randomly and at least once a hour, I had one attempt at DD that lasted all day, but froze/kp'd at night just like my original 10.6.4 that I upgraded to 10.6.5 10.6.6 via combo updates. I now leave my 10.6.4 on my ssd as a fall back that works no matter what and stays up and doesn't kp at all (In all honesty, I can make it KP if I run parrallels and try to run win7 and xp or 2 virts, or pound the virt) but with no parallels 6 running its rock solid with everything working.

 

I've taken the Firewire card out, I was also running 12 megs of OCZ pc3 12800 memory (that post always said was 1066 (even though its faster memory) and said I had 8 gigs of memory), until osx booted and told me I had the correct 12gigs, so this time I am removing half of the memory.

 

other than that, nothing except the GT8800 video card is in the machine and Im using only light blue ports (so Im going to remove ALL jmicron.kexts) and try that. I've had my firewire drives turned off as well as on and connected dasiy chain to the onboard firewire (rather than the fw800 pcie/1 card I have used forever with my 10.6.4 install).

 

I ran (optimize defaults load) on bios and changed the minimum (native ide/etc/etc) and I've loaded a minimal amount of apps to still be able to do my job (adium, office, chrome, firefox) and still cant find anything that is stable.

 

I thought maybe it was iStat, so I uninstalled that. we shall see

 

I have one question, why is it that people are using the sleepenabler kext? Sleep has always worked for me and I thought we all had the same Motherboard, so Im wondering why some people do and others dont?

 

we'll see... at least I know I can fall back to a very stable 10.6.4 but that sure would suck not being able to update anymore. What really changed in 10.6.5 anyway? I think I read a kernel change from 4 to 5 or something like that, so I know its a more major one (as compared to 10.6.1, .2, .3, .4 which didn't change the kernel, I dont know, just glad that I can boot back when I get frustrated.

 

we shall see.

Link to comment
Share on other sites

Just a follow up on my random reboot issue.

 

I cut out as many kexts as I could, and nothing changed. Today, I ran a separate cable from my power supply (modular) to the SSD, which was previously being powered by a cable that was also powering two regular hard drives. It's been several hours and everything is perfect, prime95 just keep chugging along. Guess my power supply just needed a little help. Thanks for all the help/advice!

 

One question: I cannot boot into 64-bit mode, it gives me a kernel panic mentioning com.driver.appleACPIplatform. I've had it booted in 64-bit mode in the past, but it's been a while. Any ideas? Google didn't give me anything relevant.

Link to comment
Share on other sites

I'm experiencing a strange problem and I have difficulties searching for it, so I'd be very happy if someone could share their thoughts with me on this one.

 

My fresh 10.6.6 installation worked fine until a few days ago. I have absolutely no idea what I changed (installed?) but when I put my setup to sleep and leave it for a while (a few minutes) it is not in sleep anymore. Pushing the power button will just boot normally. To me it feels like mij RAM is just walking away... One thing I suspect is my overclock so I will check that later, but it's a mild one and this system has survived more rigorous OC's before.

 

Anybody?

Link to comment
Share on other sites

:(

Hi , help please

what is this error in my kernel log?

 

Jan 30 22:20:48 Mac-Pro-di-Carmine kernel[0]: Warning - kext com.apple.iokit.CHUDKernLib has immediate dependencies on both com.apple.kernel* and com.apple.kpi.* components; use only one style.

Jan 30 22:20:51 Mac-Pro-di-Carmine kernel[0]: Warning - kext com.apple.iokit.CHUDProf has immediate dependencies on both com.apple.kernel* and com.apple.kpi.*

 

Best regard .

Link to comment
Share on other sites

I'm experiencing a strange problem and I have difficulties searching for it, so I'd be very happy if someone could share their thoughts with me on this one.

 

My fresh 10.6.6 installation worked fine until a few days ago. I have absolutely no idea what I changed (installed?) but when I put my setup to sleep and leave it for a while (a few hours) it is not in sleep anymore. Pushing the power button will just boot normally. To me it feels like mij RAM is just walking away... One thing I suspect is my overclock so I will check that later, but it's a mild one and this system has survived more rigorous OC's before.

 

Anybody?

 

I've experienced this problem before, and I believe the culprit was the overclock, somehow. I found that I could have a system that ran fine, but would not come out of sleep properly. Try running everything at stock for a while, and see if that makes a difference. Then, adjust your overclock to work.

 

It may be something other than that, I'm not expert, but that's my suggestion to you.

Link to comment
Share on other sites

Probally should have done less at one time, might make it easier to identify the issue(s), but I assume putting stuff back in one at a time will make more sense, as long as stability is reached at each step.

 

Anyways, I finally installed from a DDScript created USB source.

I also removed 6GB of memory. (BIOS post only said I had 8GB of memory, but once in OSX it showed I had 12gb like it should, not sure if that right there was the issue or not, will only be able to tell after putting it back in)

 

I also removed the PCIex1 FW800x2 FW400x1 card from the system. Connected fw devices through onboard fw at the present time.

 

I also removed all JMicron kexts from the system as I dont use those ports at all.

 

Added the sleepenabler.kext for 10.6.6 with pmversion=21. (Even though I DONT put my system to sleep, just didnt like that putting it to sleep it never woke up, which it does now.)

 

Anyways. been stable for 4-5 hours of no use, of course, the real test is leaving the office, I've yet to return to a 10.6.6 system the next morning not KP'd, but using DD's script via usb install, I had no issues at 10.6.4 over the weekend, put thats nothing new, my Kakewalk 2.2 install of 10.6.4 has been stable and up since 10.6.4 combo, so we shall see.

 

oh ya, running the RC5 mozo bootloader

 

 

Thanks for the pointers and suggestions.

 

 

 

 

 

 

 

 

 

Thanks guys, I would have never thought to try that, r115 allowed it to boot, where I was greeted with a cannot continue memory error. So I am re-making the stick and following what you said exact, copy the extra, update with r115, allow to recreate extra, and then erasing extra and restoring the original from step 15's creation.

 

Between then and now, I have been using my stable 10.6.4 (kakewalk) install to try running the script and install to another harddrive in the system. No matter what I try or who I use (tonymac's [url="http://www.insanelymac.com/forum/topic/279450-why-insanelymac-does-not-support-tonymacx86/"]#####[/url] method, DD's script with chameleon as well as pci-efi 10.5/6, Kakewalk 3.1) they all do the same thing, although most of them actually KP randomly and at least once a hour, I had one attempt at DD that lasted all day, but froze/kp'd at night just like my original 10.6.4 that I upgraded to 10.6.5 10.6.6 via combo updates. I now leave my 10.6.4 on my ssd as a fall back that works no matter what and stays up and doesn't kp at all (In all honesty, I can make it KP if I run parrallels and try to run win7 and xp or 2 virts, or pound the virt) but with no parallels 6 running its rock solid with everything working.

 

I've taken the Firewire card out, I was also running 12 megs of OCZ pc3 12800 memory (that post always said was 1066 (even though its faster memory) and said I had 8 gigs of memory), until osx booted and told me I had the correct 12gigs, so this time I am removing half of the memory.

 

other than that, nothing except the GT8800 video card is in the machine and Im using only light blue ports (so Im going to remove ALL jmicron.kexts) and try that. I've had my firewire drives turned off as well as on and connected dasiy chain to the onboard firewire (rather than the fw800 pcie/1 card I have used forever with my 10.6.4 install).

 

I ran (optimize defaults load) on bios and changed the minimum (native ide/etc/etc) and I've loaded a minimal amount of apps to still be able to do my job (adium, office, chrome, firefox) and still cant find anything that is stable.

 

I thought maybe it was iStat, so I uninstalled that. we shall see

 

I have one question, why is it that people are using the sleepenabler kext? Sleep has always worked for me and I thought we all had the same Motherboard, so Im wondering why some people do and others dont?

 

we'll see... at least I know I can fall back to a very stable 10.6.4 but that sure would suck not being able to update anymore. What really changed in 10.6.5 anyway? I think I read a kernel change from 4 to 5 or something like that, so I know its a more major one (as compared to 10.6.1, .2, .3, .4 which didn't change the kernel, I dont know, just glad that I can boot back when I get frustrated.

 

we shall see.

Link to comment
Share on other sites

I've experienced this problem before, and I believe the culprit was the overclock, somehow. I found that I could have a system that ran fine, but would not come out of sleep properly. Try running everything at stock for a while, and see if that makes a difference. Then, adjust your overclock to work.

 

It may be something other than that, I'm not expert, but that's my suggestion to you.

You were right and as I myself also expected, it was the overclock. I took it off and sleep worked right away. The problem is that my motherboard squeals when it runs at stock, this seems to be a common problem.

Link to comment
Share on other sites

You were right and as I myself also expected, it was the overclock. I took it off and sleep worked right away. The problem is that my motherboard squeals when it runs at stock, this seems to be a common problem.

Hmm, overclocking shouldnt really affect your sleep function so I find it very weird that yours does. I can overclock to 4.2 Ghz and everything works flawlessly. My MB has never squealed before whether at stock speed or overclocked so something must be wrong with yours.

Link to comment
Share on other sites

I'll give it another try. But, please tell me how do i add this "EFI string" ?

Where do i have to add this ? On the Empire EFI boot CD ?

Please, don't mind me, Digital_Dreamer, but i really have no idea.

:) Thanks again.

PS: i'm close, i feel that :D I will use SL, very soon.

Help me, please.

Link to comment
Share on other sites

Hmm, overclocking shouldnt really affect your sleep function so I find it very weird that yours does. I can overclock to 4.2 Ghz and everything works flawlessly. My MB has never squealed before whether at stock speed or overclocked so something must be wrong with yours.

 

Well, that makes two people though that have had the overclock affect sleep, so I guess I'm not crazy. I tried upping my QPI voltage and it helped some, although it still has issues occasionally. I'm a novice overclocker though, so I could have messed things up somewhere.

Link to comment
Share on other sites

I've been following the post's reply's for the last couple of weeks, and even started reading from where I left off back when I installed 10.6. So I apologize if its been asked and my search choice/combo of words isn't flagging it. But here is my question, as I start to add my pulled hardware back into the machine to test each piece, should I be re-running the scripts DDST routines? I would assume that a card being added, or memory being added I should be re-running it?

 

I also noticed during a verbose boot (and remembered that It was doing it before but I gave up on worrying about it because it didn't affect anything), but with the pci/e firewire card I get LOTS of these in my log files

 

 

2/1/11 1:09:33 PM kernel FireWire GUID 0000000000000000 is invalid!

 

 

 

Also, one guy back was asking how I put this card into my machine because on his pcie/1 slot it hit the cooler die, I just realized that even though its a pcie/1 lane card, I have it in the pcie/4 lane slot. Hope that helps.

Link to comment
Share on other sites

Hmm, overclocking shouldnt really affect your sleep function so I find it very weird that yours does. I can overclock to 4.2 Ghz and everything works flawlessly. My MB has never squealed before whether at stock speed or overclocked so something must be wrong with yours.

 

 

Well, that makes two people though that have had the overclock affect sleep, so I guess I'm not crazy. I tried upping my QPI voltage and it helped some, although it still has issues occasionally. I'm a novice overclocker though, so I could have messed things up somewhere.

 

My overclock was very quick and probably was very unstable this time. Normally I spend a bit more time getting things stable but I suppose I was just a bit too inpatient.

 

It didn't make sense to me either, but as it turns out, going back to stock speeds solved the problem. I think that the OC affected my (simple Kingston) RAM and since this is used during sleep it caused the problem.

 

With regards to the squealing, it is something that has been there from the start and I'm ok with it. I decided not to go through the hassle of RMA with Gigabyte, even though it is probably part of warranty.

 

An odd thing happened yesterday when I put it to sleep: the fans spun up and the display went off like normally, but the system didn't power down. So the fans kept spun up. It might have been a once-only hickup but I'll keep an eye on it.

Link to comment
Share on other sites

I re-ran the DSDT dump (saving my old DSDT), ran a diff against them and they are different, so obviously it makes sense to re-run it as you add or subtract software? yes?

 

Anyways, I did some searching for the GUID 00000000 definition on the gigabyte dsdt thread, but my dsdt contains nothing like they have you looking for, not sure if thats the issue, (that its not mentioned) and thus why it gets the 000000000 guid, here is it getting recognized in dmesg if anyone can help with stopping this, it adds to my console ever 5 seconds, and makes it a huge file..

 

<from my dmesg>

 

IOBlockStorageDriver/INTEL SSDSA2M080G2GC Media/IOGUIDPartitionScheme/SSD-osX@2

BSD root: disk0s2, major 14, minor 3

FireWire (OHCI) TI ID 823f PCI now active, GUID 0000000000000000; max speed s800.

RealtekRTL81xxx.kext: NIC identified as RTL8168D/8111D (mcfg=18)RealtekRTL81xxx.kext: NIC identified as RTL8168D/8111D (mcfg=18)

 

 

when I pull the add-on Firewire board and turn on the motherboards fw 400, the OHCI line above reads ->

 

FireWire (OHCI) TI ID 8024 PCI now active, GUID 007bbc070000241d; max speed s400

 

Im sure I need to add a GUID to it somehow,

 

thanks in advance.

 

SD

 

 

 

 

 

 

I've been following the post's reply's for the last couple of weeks, and even started reading from where I left off back when I installed 10.6. So I apologize if its been asked and my search choice/combo of words isn't flagging it. But here is my question, as I start to add my pulled hardware back into the machine to test each piece, should I be re-running the scripts DDST routines? I would assume that a card being added, or memory being added I should be re-running it?

 

I also noticed during a verbose boot (and remembered that It was doing it before but I gave up on worrying about it because it didn't affect anything), but with the pci/e firewire card I get LOTS of these in my log files

 

 

2/1/11 1:09:33 PM kernel FireWire GUID 0000000000000000 is invalid!

 

 

 

Also, one guy back was asking how I put this card into my machine because on his pcie/1 slot it hit the cooler die, I just realized that even though its a pcie/1 lane card, I have it in the pcie/4 lane slot. Hope that helps.

Link to comment
Share on other sites

 Share

×
×
  • Create New...