Jump to content

[GUIDE] Scripted Yosemite/Mavericks Install on Gigabyte Mobos


4,696 posts in this topic

Recommended Posts

So far no KP, crossing fingers ... That last posting was an anomaly, lucky to have it stable for the post.

 

Your suggestion to check the Micron vs Intel was a good one. I did have to cannibalize my phantom Green drive a few weeks ago (external switch went bad after a year - boo). It was hooked in correctly (Gold drive) but, I inadvertently had the Seagate external eSata drive on the JMicron controller. It was probably messing up the whole bot? Still no KP so perhaps it worked. Sure appreciate the tip.

 

Sorry to not understand better but, can you say more please about your suggested -ls prompt?

 

Try literal: rm -rf /Volumes/FAKE_MOUNT_POINT

Try partial: rm -rf /Volumes

 

No go for either command set.

 

OMG - just googled this, yo must be suggesting I remove the extra files if I See them? I thought you were telling me how to see them. Hope I didn't just make it worse. Still, no KPS though, perhaps out of that trap at least (whew.

 

 

ADDED: Stll no KPs, whew. I think I can work through this. I htink the ls prompt was not working becasue UI did not realize I needed to go CD to the main directory first. Doing this makes sense now from what you said. TO be as clear as possible (especailly when about to delete something) I show:

 

From LS:

 

Gold Drive Patriot

Macstart1 RAID Drive

Macstart2 Seagate External

Name_of_your_install_volume

 

Is the term " Name_of_your_install_volume" actually a bogus phantom term that will be safe to delete?

 

Example: rm -rf /Volumes/Name_of_your_install_volume

 

Correct?

 

Thanks

 

ZoroLives,

I'm assuming your RAID drives are on the JMicron controller. If so, I may suggest moving them to the Intel controller, at least temporarily, and see if it resolves your KPs. Worth a try.

 

As for the duplicate mount points:

The mount list you posted should be find. They are what's actually mounted. The issue is how it compares to the ls command in Terminal, as those are what the system uses as mount points. If you have a "drive" that shows up in ls that isn't in the mount list, that would be a problem and would need to be deleted once you are sure it is not a valid drive. Just do rm -rf /Volumes/FAKE_MOUNT_POINT

 

Provide that list here, if you can.

 

best of wishes,

MAJ

Link to comment
Share on other sites

So far no KP, crossing fingers ... That last posting was an anomaly, lucky to have it stable for the post.

Good to hear! That would be aggravating.

 

Sorry to not understand better but, can you say more please about your suggested -ls prompt?

Sorry, I didn't make this clear - I reread and realized I left out some info - my fault. :)

Just open the Terminal and type ls /Volumes (lower case "L") and it'll list the mount points in that directory.

The /Volumes directory contains the mount points for all of the drives (other than the boot volume) connected to the system. To put it simply, it just lists the mounted volumes and you can access all files from these volumes from this point (e.g. /Volumes/HD_DRIVE/Some_File). The Finder hides the Volumes directory itself, but displays its contents at the Computer level.

 

List the contents of ls here for us to see.

If it contains something other than what you really have mounted, then it may be false and can cause confusion for software/scripts that refer to it. The script may check to see if a partition is mounted and see that FAKE_VOL is listed in /Volumes and think it is mounted when it really isn't.

 

Extra Info for the curious:

This can happen if one is writing to a partition (via script) and, for some reason, the script is canceled/abruptly quit/something goes wrong/etc. and the partition gets unmounted while the script is still writing files to it. The OS is supposed to clear away any unmounted volumes listed in /Volumes directory and actually does so, but if the script is still writing files to that directory (/Volumes/PREVIOUSLY_MOUNTED_VOLUME/somefiles), it will actually build up a "volume" containing the files the script is writing, even though that "volume" is unmounted. So, the files are no longer on the real volume, but, instead, stored in the /Volumes directory.

 

Now, if one were to look into /Volumes, they will see /Volumes/PREVIOUSLY_MOUNTED_VOLUME/somefiles, even though that partition/volume has been unmounted. Hence, it's fake or false, and can trick a script into writing to it, instead of actually remounting the correct partition/volume and writing to it.

 

Of course, if the script were to remount the partition/volume, that partition/volume would gain a suffix PREVIOUSLY_MOUNTED_VOLUME 2 (because it shares the same name as the other) and still wouldn't be written to, because it's not the correct name/path. Hence, the script will continue writing to the false mount point.

 

This is why I take all this rather seriously and make a big deal out of it. It can cause a lot to go wrong rather quickly.

 

Just for anyone's information - I haven't had a issue with false mount points in the last six months, in spite of the fact that I try testing the script in all sorts of scenarios. I will periodically force a false mount point for the purpose of testing, however.

 

MAJ

Link to comment
Share on other sites

Mine could have stemmed from my (formerly) unstable JMicron attachment. I have posted all the info I think, here it is all together:

 

Mac-Pro:~ myuser$ mount -uw

/dev/disk2 on / (hfs, local, journaled)

devfs on /dev (devfs, local)

fdesc on /dev (fdesc, union)

map -hosts on /net (autofs, automounted)

map auto_home on /home (autofs, automounted)

/dev/disk3s2 on /Volumes/Gold Drive (hfs, local, journaled)

/dev/disk1s2 on /Volumes/Macstart1 (hfs, local, journaled)

/dev/disk0s2 on /Volumes/Macstart2 (hfs, local, journaled)

/dev/disk4s3 on /Volumes/Patriot (hfs, local, nodev, nosuid, journaled, noowners)

 

Mac-Pro:~ myuser $ ls /Volumes

Gold Drive

Macstart2

Patriot

Macstart1

Name_of_your_install_volume

RAID Drive

 

Is the phrase, "Name_of_your_install_volume" a bogues placeholder/ghost/non existant volume that was left over?

 

Much thanks.

 

ADDED: >>--->

I can't say enough how grateful I am for all your hard work on this script and how much you have helped me. In case it is useful to say it, this is what I find confusing about the script error message and why I would still appreciate some confirmation that I am about to delete the correct data.

 

Script error message:

There appears to be duplicate mount points.

 

If Mac OS X is restarted unexpectedly, the current mount point listed

in /Volumes may not be deleted and a new ("false") mount point with a

"1" appended to the name may appear (such as "My Volume 1"). This can

also happen when a script is writing files to an unmounted drive. This

can cause issues for applications and scripts that rely on the path to

certain files located in secondary volumes.

 

NOTE: The "false" mount point may or may not be the volume whose name

ends with "1". In fact, it may not even have a number appended to the

name, if the other volume has become unmounted. In the Finder, use Get

Info (or on the command line, use du -sk) to determine which mount point

is smaller in size -- the smaller one is likely the "false" share point.

 

If you are confident that the listed mount point is, indeed, a duplicate,

is a false volume, and is not mounted, you can delete it in Terminal:

sudo rm -r "Volumes/false_mount_point_name"

 

SEE APPLE SUPPORT PAGE: (http://support.apple.com/kb/TA24241)

 

The suspect mount points on your system are as follows:

"Name_of_your_install_volume"

 

I wish to resolve this issue before continuing [Y/n]:

When you say:

If Mac OS X is restarted unexpectedly, the current mount point listed

in /Volumes may not be deleted and a new ("false") mount point with a

"1" appended to the name may appear (such as "My Volume 1").

and

you can delete it in Terminal:

sudo rm -r "Volumes/false_mount_point_name"

I start looking for the drives actual name as in "Gold" or "Gold 1" as a duplicate mount point. But it appears that when you say:

The suspect mount points on your system are as follows:

"Name_of_your_install_volume"

You mean literally that there could be a string named "Name_of_your_install_volume". It has seemed confusing as I have been reading this error message as to whether I should be looking for a literal ("Name_of_your_install_volume") or a representative ("Gold 1") phrase.

 

Ongoing thanks.

 

 

Good to hear! That would be aggravating.

 

Sorry, I didn't make this clear - I reread and realized I left out some info - my fault. :)

Just open the Terminal and type ls /Volumes (lower case "L") and it'll list the mount points in that directory.

The /Volumes directory contains the mount points for all of the drives (other than the boot volume) connected to the system. To put it simply, it just lists the mounted volumes and you can access all files from these volumes from this point (e.g. /Volumes/HD_DRIVE/Some_File). The Finder hides the Volumes directory itself, but displays its contents at the Computer level.

 

List the contents of ls here for us to see.

If it contains something other than what you really have mounted, then it may be false and can cause confusion for software/scripts that refer to it. The script may check to see if a partition is mounted and see that FAKE_VOL is listed in /Volumes and think it is mounted when it really isn't.

 

Extra Info for the curious:

This can happen if one is writing to a partition (via script) and, for some reason, the script is canceled/abruptly quit/something goes wrong/etc. and the partition gets unmounted while the script is still writing files to it. The OS is supposed to clear away any unmounted volumes listed in /Volumes directory and actually does so, but if the script is still writing files to that directory (/Volumes/PREVIOUSLY_MOUNTED_VOLUME/somefiles), it will actually build up a "volume" containing the files the script is writing, even though that "volume" is unmounted. So, the files are no longer on the real volume, but, instead, stored in the /Volumes directory.

 

Now, if one were to look into /Volumes, they will see /Volumes/PREVIOUSLY_MOUNTED_VOLUME/somefiles, even though that partition/volume has been unmounted. Hence, it's fake or false, and can trick a script into writing to it, instead of actually remounting the correct partition/volume and writing to it.

 

Of course, if the script were to remount the partition/volume, that partition/volume would gain a suffix PREVIOUSLY_MOUNTED_VOLUME 2 (because it shares the same name as the other) and still wouldn't be written to, because it's not the correct name/path. Hence, the script will continue writing to the false mount point.

 

This is why I take all this rather seriously and make a big deal out of it. It can cause a lot to go wrong rather quickly.

 

Just for anyone's information - I haven't had a issue with false mount points in the last six months, in spite of the fact that I try testing the script in all sorts of scenarios. I will periodically force a false mount point for the purpose of testing, however.

 

MAJ

Link to comment
Share on other sites

Guys I've just upgraded my machine to 10.6.5 using the apple software update and it's killed various things.

 

Nothing adobe Flash related seems to work at all (so thats all my browsers except Safari, and Safari is very Flash-unstable). I can't run the adobe flash install manager which is annoying. I also don't know how to uninstall the current version of Flash and reinstall a different version.

 

iTunes also doesnt work. So maybe Flash isnt the only problem. Error meesage is attached hopefully.

 

Things like Reaper and Matlab still seem to work ok though.

My original install was 10.6.2 and i've just automatically stepped through to 10.6.5 via the other two updates without too many problems (although flash was becoming more unstable in 10.6.4).

 

My machine is a UD5 with 9400 passive graphics card and some other stuff.

 

HELP!!! Matt :D

 

As an aside: About this Mac has always said its a Quad Core Intel Xeon although the OS does show 8 CPUs on boot up. How can I view how much each core is being used and whether they are all being used? (Probly a daft question!)

chrome_crash.rtf

Link to comment
Share on other sites

I wish you well with that. I am no longer able to boot. I had to go to a friends house and am on his Mac. Will have to start at the bottom again, build a boot disk (which I should have had anyway).

 

Last time I ran the scripts it would not see the graphics card. So I got out the old EFIStudio 1.1 and I guess I should not have done it, not it's all gone. Que Bummer.

Link to comment
Share on other sites

ZoroLives,

Thanks for that post.

 

Is the phrase, "Name_of_your_install_volume" a bogues placeholder/ghost/non existant volume that was left over?

LOL. I see the confusion!

No, it's not a placeholder, but the actual name of the "volume" or mount point.

I don't know where it came from, but I can certainly say it didn't come from my script, as that phrase is nowhere in the script. Additionally, all my variables in the script are in all-caps, as well.

 

From the postings, I would say both "Name_of_your_install_volume" and "RAID Drive" are false mount points and need to be deleted. However, I would also say that those false mount points would not create a problem until your system attempts to mount or work with a volume having the exact same name. So, if you have a real volume called, "RAID Drive", then you'll have problems.

 

Those names are not coming from the script. The ones I usually worry about are "EFI" for EFI setups (all GUID volumes share the same EFI name) and "Boot OSX" from Apple RAID setups (by default, Apple RAID helper partitions share this same name, until they are changed by the user).

 

You mean literally that there could be a string named "Name_of_your_install_volume". It has seemed confusing as I have been reading this error message as to whether I should be looking for a literal ("Name_of_your_install_volume") or a representative ("Gold 1") phrase.

I totally understand what you're saying and the source of the confusion. That was an odd "Volume" name and, yes, in this location, it's literal, not a placeholder. However, it was a placeholder when it was issued in the Terminal at one time. Likely, a script or series of Terminal commands were issued and that placeholder was not renamed to the desired volume name.

If you use the Terminal to copy files to a volume that is non-existent, this can happen. Say, you issue cp THIS_DIRECTORY /Volumes/NON_EXISTENT_VOLUME, those files may get copied and create a false mount point in /Volumes/NON_EXISTENT_VOLUME, even though there is no volume by that name. That's why it's important for scripts or Terminal users to verify or ensure the presence of the volume in use.

 

Now, to determine if they REALLY are false:

1. If you don't have a volume/partition using the name displayed, delete it.

NOTE: There are volumes a user may never normally see, so beware. Those are just "EFI" partitions and Apple RAID helper partitions, "BOOT OSX", by default (slices don't have names).

2. Just check the size of the alleged false mount point (in Terminal). False mount point "volumes" are typically very small in size (or even zero, if it just has directories/folders), depending on what files got copied over during the mishap.

 

If you have a volume named, "MY_DRIVE" and there are two listed in /Volumes, with one having a number suffix added ("MY_DRIVE" and "MY_DRIVE 2"), which one is the real volume and which is fake? Again, the answer is the one that's smaller in size.

Issue the following command in Terminal:

du -sh /Volumes/FAKE_VOLUME

If the size is really small, like a few kilobytes, and you know you have 4GB of capacity in use on that drive, then this is the false mount point.

 

best of wishes,

MAJ

 

i have another boot disc (10.6.2) but for some reason when i select it in the bootloader i always end of booting off my normal 10.6.5 install.

 

Any ideas how I get around this? It's ages since I built the discs and I've kinda forgotten stuff...

Your boot drive (the one given priority in the BIOS), which I guess is the 10.6.5 drive and not the one you want to boot, probably has a "boot-uuid" flag in the boot.plist that points to the 10.6.5 drive, thus forcing it to always boot that drive.

 

You can remove that flag, if it's not a EFI setup. A EFI setup needs that flag to know where the system partition is located.

Or, you can change the BIOS drive order.

 

Hope that helps.

MAJ

Link to comment
Share on other sites

Thanks. I was able to haul out the old G4 and burn a Kalyway DVD which as it turns out will allow my old RAID drive to boot so long as it is in the optical drive. It isn't actually booting to install a fresh system, it is just allowing the old boot up to occur.

 

My Goal: To get this Raid setup updated and make it my main boot disk again. Last December I started having some very tough problems with my system. I moved to my old backup and made that my main drive. Now this drive, which is the only one that I can still boot to is the old Raid drive with 10.5.

 

My current main drive will not boot anymore, it has been updated to 10.6.4 from 10.6.2. It is also backed up to a Seagate external drive as 10.6.2.

 

Problem: I cannot get any combination of script to work so far in any way to get either of my 10.6 drives working. I need to do that to get back online, missed a good bit of business this week so far. Once I get booted to my main drives, I want to go after this Raid setup and make it current and understand how to maintain it better.

 

Question one would be, what is the recommended meathod for installing a bootable disk via the script?

 

Observation: It would be helpful if all the different versions of the boot loaders had a help file I could see to explain why I should or should not use some of those boot schema (customized as mozo r642, pre12 r115, RC4 r684, RC3 r685, Asere (Unofficial...) It would reduce my cycles of trail and error if I knew that some are just not set up for this system.

Current Chameleon bootloaders available to install:

1) Skip or cancel

2) Cham 2.0 RC5 mozo (r642 - Unofficial) Based on modifications to RC4

3) Cham 2.0 RC5 pre12 (r115 - Unofficial) Based on modifications to RC4

4) Cham 2.0 RC4 Asere (Unofficial) Based on AsereBLN 1.1.9 and PC-EFI v10.5

5) Chameleon 2.0 RC4 (r684) Latest official release from the Chameleon team

6) Chameleon 2.0 RC3 (r658) Previous official release from Chameleon team

7) PC-EFI v10.6 Latest version from netkas. Based on Chameleon RC3

8) PC-EFI v10.5 Previous version from netkas. Based on Chameleon RC3

9) Chameleon v1.0.12 Old, non-GUI version. Incompatible with vanilla kernel.

Much thanks for your help (anyone). I know I am not the only one here needing assistance.

Link to comment
Share on other sites

Hi Zorolives.

 

If you really need to get things up and running and your missing out on business wouldn't the easiest solution be to purchase a new HD and do a fresh install onto that. I presume you had a working install previously so you could just do what ever you did to get to that point on the new HD then access your Apple RAID stuff as non-booted drives.

 

Once you have that up and running you could copy across anything you needed then work fresh on the RAID and copy everything back once your done.

 

I'm also running an Apple Software RAID and playing around with different bootloaders and kexts but I do all my playing around on a separate HD until I find something solid that works.

Link to comment
Share on other sites

Yes, I may have to do that. I was just trying out an old IDE drive for that purpose. Turns out I was getting the old Panic so that is unhooked. I was at 10.6.2, watching others challenges with 10.6.3, & 4. 10.6.4 sounded more stable and I took the plunge. BUT!!! 10.6.5 came out while I have been trying all this out. ARRRGH!!! Now my two main drives are 10.6.5 and most of the hack tips are not yet confident on that kernel version.

 

At least I am currently booting on the one RAID volume (on and off again).

 

Keeping at it.

 

Thanks

 

 

Hi Zorolives.

 

If you really need to get things up and running and your missing out on business wouldn't the easiest solution be to purchase a new HD and do a fresh install onto that. I presume you had a working install previously so you could just do what ever you did to get to that point on the new HD then access your Apple RAID stuff as non-booted drives.

 

Once you have that up and running you could copy across anything you needed then work fresh on the RAID and copy everything back once your done.

 

I'm also running an Apple Software RAID and playing around with different bootloaders and kexts but I do all my playing around on a separate HD until I find something solid that works.

Link to comment
Share on other sites

Just a heads up.

 

I tried out the included sleepenabler for 10.6.5 but it was giving me KP while trying to boot. I did some searching and found that alot of people had problems with the initial 10.6.5 sleepenabler while using 64 bit boot so I suspect this is the cause of the problem.

 

I found a sleepenabler that has been modified to work in 64 and 32 bit and it seems to be working fine.

Link to comment
Share on other sites

Thanks, I appreciate the suggestion. I already found that, it did not help. There are so many options spread out

through so many threads and looked at on more than a few hackintosh oriented sites. I need to understand the

basics. I apologize but it seems I have not yet fully grasped what needs to be where to boot - or if we are saying

the same thing when we say "to boot".

 

First, using a myriad of boot loaders out there I am able to boot to where I can see all my drives. Then I can

arrow down and boot single user, boot ignore caches, etc. If this means booting, then I have done it. The trouble I

have is what happens after I try a selection. I have tried all phased of the DD script, and every other GUI

interface I could find across the net for installing OSX on PC hardware.

 

I have gotten results as: Sudden restarting, sudden KP with small grey multi language rectangle in the center,

lines and lines of verbose code then stopping at a kp in the register (not a frozen screen, just a halted script.

Apple logo and spinning "working" synmbol, going on for hours lines and lines of cache rebuildign going on for

hours.

 

I have treid past sucesses which are not working now such as booting on one sucessful method then at seeing the

disk prompts, moving to another disk and trying to continue to boot.

 

I need to handle this, step by step, manually. I keep throwing the dice at the problem and that is just not

working. There are just too many variables and I am already starting to use my older semi-stable drive just to get

back to some work things. But I really need to get back up and running. I need to see it in one place, there have

been a few discussion threads about what does and does not work with 10.6.3, 10.6.4, and now 10.6.5. I "was" stable

(for the most part) on 10.6.2. I heard sleepenabler was no longer needed, now it appears it is still needed but

needs to be a newer version. I will start culling the threads to se if I Can find what I need. If anyone has

already found this information, what ALL is needed to succesfully noot 10.6.5 I would be very grateful.

 

Please note, some will say, "It worked for me exept I had to do somethig to fix audio or some such". This is fine

for what it is but this is not helpful to me. I am asking so that I will understand what a succesful install

bootloader would look like. I Need to see what pieces are mising or in the wrong place or not updated yet

(speepenabler). I also have read that 10.6.5 requires a new kernal.

 

ADDED: I am wondering what will happen if I just re-install SNL on one of my drives I am trying to boot too? I don't

want to wipe the drive so will doing this wipe my drive? I am assuming that most of my program serial numbers are

stored along with email in the user system folder. If true then I assume re-installing a system again would not

hurt?

 

I hope that I can get this going again, and soon.

 

Much thanks for any assistance.

 

Just a heads up.

 

I tried out the included sleepenabler for 10.6.5 but it was giving me KP while trying to boot. I did some

searching and found that alot of people had problems with the initial 10.6.5 sleepenabler while using 64 bit boot

so I suspect this is the cause of the problem.

 

I found a sleepenabler that has been modified to work in 64 and 32 bit and it seems to be working fine.

Link to comment
Share on other sites

i have another boot disc (10.6.2) but for some reason when i select it in the bootloader i always end of booting off my normal 10.6.5 install.

 

Any ideas how I get around this? It's ages since I built the discs and I've kinda forgotten stuff...

Press the delete key below the help key to go into the bios at startup and switch the boot priority to your 10.6.2 drive.

 

Zorolives:

Are you trying to run the script from a 10.5 boot? I remember there was some additional stuff you needed to do to make it work but I don't remember the details since it was a long time ago for me. If you are then I think there is a help guide for this stuff on the first page.

 

In order to get things running I'd suggest trying to use the PCEFI 10.6 bootloader with as little kexts as possible (Probably just FakeSMC and NullCPUManagement). I'm not sure if you'll need a kext for your graphics but as a guess I'd say PCEFI 10.6 can probably handle it on its own but I think you need to boot into 64 bit for that so delete the arch=i386 plist.

Link to comment
Share on other sites

I used this excellent guide back in Nov 09, shortly after it was posted, to install SL on my then new water cooled overclocked hardware - it worked like a dream and has been a very reliable setup ever since.

 

Stuck at home today by the heavy snow I decided to update everything and ended up installing a brand new SL on a 4x750Gb RAID0 setup.

 

Once again your script worked perfectly and I was up and running in no time - man does it boot up and run fast!

 

I just wanted to say thank you - clearly, a considerable amount of time and effort has gone into this script.

 

Thank you! Keep up the good work.

 

Kev

Link to comment
Share on other sites

I wish i could install Snow Leopard, too, on my PC.

Gigabyte EX58-UD5, Intel 7 920, 3 GB DDR3, 2 X WD 6400 AAKS.

Did you install SL over another version of MAC OS X ?

All i can install is 10.5.7 but i can't find the drivers.

Please, someone explain to me hiw do i have to proceed, to get SL.

Thanks.

Link to comment
Share on other sites

I wish i could install Snow Leopard, too, on my PC.

Gigabyte EX58-UD5, Intel 7 920, 3 GB DDR3, 2 X WD 6400 AAKS.

Did you install SL over another version of MAC OS X ?

All i can install is 10.5.7 but i can't find the drivers.

Please, someone explain to me hiw do i have to proceed, to get SL.

Thanks.

 

 

Hey man , Did you really read this Guide ?

 

:unsure:

 

Come on !!

 

Many people here have spend weeks,days and hours to make this guide and lots succeeded !

 

So first, a little effort to read.....................

 

after that you can ask for particular help

 

 

Successs

 

:)

Link to comment
Share on other sites

You didn't understand me.

I wanted to ask the way i have to proceed. Do i need SL DVD retail ?

Ok, if so, do i have to add some files on this DVD SL image ?

I tried to understand this guide, but is very hard. My english is bad.

Link to comment
Share on other sites

OK, back again (almost). I really need to put together a how-to guide at least for myself for the next time. I have the orange icons for drives. ThE spotlight index seems to be a no show. Yes I have used Terminal to activate it and remove it and rebuilt it, it still isn't functioning for some reason. I also had that 6 gig installed memory bit it sees only 4 gigs thing.

 

All I did was use [url="http://www.insanelymac.com/forum/topic/279450-why-insanelymac-does-not-support-tonymacx86/"]#####[/url] or other boot iso files you can use to burn a bootable disk, swap in the retail SNL install disk, and reinstall 10.6.0 on my old main drives. (NOTE TO THE OTHER RECENT POST ABOUT HOW TO GET STARTED: THAT IS HOW IT WORKS) Now of course I still need to remember, what goes where. Darn it, I am still confused about that. I would like to migrate up to 10.6.5 but for now I have gotten so behind on work I Want to get stable again at 10.6.2. I ran the 10.6.2 updater so that is where I sit for now.

 

I know the terms are sprinkled through out this and other threads. What I would like to see, and maybe I Can do it and give back to the community if it would help anyone else, would be to see, visually, a map of what parts have to go where for a successful boot. I cant take the time to do that with many configs, but it might help me as a recovering noob, as well as other noobs. It might also give cause for the experts around here to take less aspirins?

 

For one, I am confused about the concept of using the extra EFI partition or PCI as reflected in the two choices in the MAJ scrip. Why is one a better choice in some circumstances, or does it matter at all? For another, it looks like there are more than one kernels, shouldn't there be just one kernel and what kernel versions will work and which ones are working for some systems but not others, or is that just the way kernels are - no one kernel will work across the board?

 

As you can see, I am more trying to understand the concepts here. As I do, I start to understand why and then I make smarter choices as to what is missing, what to try and what not to experiment with at all. As it is, I have gone through the script on this thread so many times and tried it so many ways. But I still don't have a clear understanding of just what I must have, and where I must have it for this to all work.

 

I know I can get verbose and I apologize for offering such a long post.

Link to comment
Share on other sites

Hi Digital_Dreamer

 

Here i'm again with a new possible bug found on your latest script. The routine that creates de Preboot CD has some kind of bug in it, cause the CD created appears as empty, and if you are using a DVD+RW and erase it using fast erase, the DVD becomes difficult to read on any DVD Drive. I tried both writing the disc from your script and writing the ISO file with disc utility.

 

The Preboot_CD.ISO that you have in your web server works greats, but it lacks from my custom dsdt.aml

 

Also, i've found that the 8) View Partitions option sometimes mounts the 2 partitions that make my RAID volume, and sometimes it leaves one of them mounted. Just FYI

 

Finally, i tested the Chameleon 2.0 RC5 Included within your script and everytime that your start up the computer says: couldn't find Default\Background.png, and you don't have a GUI for chameleon, just text version of it.

Link to comment
Share on other sites

Finally, i tested the Chameleon 2.0 RC5 Included within your script and everytime that your start up the computer says: couldn't find Default\Background.png, and you don't have a GUI for chameleon, just text version of it.

 

Yeah I got this bug too but its easily fixable. Just copy the background.png file from one of the other themes and u will start with the GUI enabled. My system has been acting so so weird after testing the new script. I thought it was because of the 10.6.5 update but I am not sure as prior to testing the script, all was well.

 

I think the script is full proof, does what its supposed to do. If anyone has some pointers or any weird behavior, please let us know. Thia is really eating me up :angel: . I hardly shut down my box, I let it sleep and pick up from where I left off. Now when I let it sleep, it just restarts a few minutes after waking up. I have done everything I know...I will test a new drive later and confirm.

 

Cheers

Charles

Link to comment
Share on other sites

You didn't understand me.

I wanted to ask the way i have to proceed. Do i need SL DVD retail ?

Ok, if so, do i have to add some files on this DVD SL image ?

I tried to understand this guide, but is very hard. My english is bad.

Yes, you need a retail SL DVD.

And, no, no additional files need to be added to that.

 

ThE spotlight index seems to be a no show.

Have you tried adding your hard drive to the Spotlight/Privacy window, closing Preferences, then opening and removing them?

 

a map of what parts have to go where for a successful boot.

I may be misunderstanding you, but modified kexts only go in /Extra or /System, and most will work in either place.

 

For one, I am confused about the concept of using the extra EFI partition or PCI as reflected in the two choices in the MAJ scrip. Why is one a better choice in some circumstances, or does it matter at all?

Simply put: it doesn't matter. It's just a matter of preference for some. I find the /Extra setup (as opposed to EFI) to be much simpler and easier to work with personally.

One is hidden and the other isn't.

 

For another, it looks like there are more than one kernels, shouldn't there be just one kernel and what kernel versions will work and which ones are working for some systems but not others, or is that just the way kernels are - no one kernel will work across the board?

If you have a Core i7 processor based system (and I see that you do), then the vanilla kernel works fine. The other kernels are for those running AMD boards, ATOM boards, or an older version of Leopard.

 

As you can see, I am more trying to understand the concepts here. As I do, I start to understand why and then I make smarter choices as to what is missing, what to try and what not to experiment with at all.

No problem at all. That's what we all are doing and we are all at different stages of this learning process.

 

The routine that creates de Preboot CD has some kind of bug in it, cause the CD created appears as empty, and if you are using a DVD+RW and erase it using fast erase, the DVD becomes difficult to read on any DVD Drive. I tried both writing the disc from your script and writing the ISO file with disc utility.

The media you're is going to play a significant role in burning success. I haven't had any issues with CD-RW, but haven't tried DVD-RW, but expect the DVD-RW to be more difficult to work with. I'd suggest using CD-R or CD-RW - they're cheaper. :)

 

Also, i've found that the 8) View Partitions option sometimes mounts the 2 partitions that make my RAID volume, and sometimes it leaves one of them mounted. Just FYI

Yeah, I've been aware of that since the start. :) It's part of the unfinished /Extra/Extensions mode I put in. The fix will be out this weekend.

 

Finally, i tested the Chameleon 2.0 RC5 Included within your script and everytime that your start up the computer says: couldn't find Default\Background.png, and you don't have a GUI for chameleon, just text version of it.

Sorry about that. The Default theme doesn't have a background.png. I hadn't noticed this issue, because I was using other themes.

It's been fixed and will be released this weekend.

 

best regards,

MAJ

 

Charles,

Do the reboots happen after a period of inactivity?

I saw your logs, but not sure if anything was showing up right prior to the reboots. Did we see all relevant info?

 

I suppose the "autologout" key was nonexistent in your .GlobalPreferences.plist?

 

kind regards,

MAJ

Link to comment
Share on other sites

The media you're is going to play a significant role in burning success. I haven't had any issues with CD-RW, but haven't tried DVD-RW, but expect the DVD-RW to be more difficult to work with. I'd suggest using CD-R or CD-RW - they're cheaper. :hysterical: MAJ

 

Hi Digital_Dreamer, i will try following your instructions, but believe me, if i burn the preboot ISO located in your server, it works with the same DVD-RW, and when using the old script 4.56 worked fine too, but not with the 5.00 beta.

 

I have a very important Question, i don't know if you are aware of this, but if you are installing SL Server, if there is no Network Card recognized by the system, then it will refuse the Serial Number when finishing setup after reboot. I remember that i solved it by connecting the HD to another Mac and applying the Script there. (I installed using the preboot method, cause is the best option)

 

My question is: Is it possible to drop some kexts at the Bootcd\extra folder of your script so when you make the preboot_cd those devices get recognized?

 

Thanks in advance.

Link to comment
Share on other sites

Have you tried adding your hard drive to the Spotlight/Privacy window, closing Preferences, then opening and removing them?

I am now trying this, since you suggested it. I have tried trashing the spotlight index in Terminal and rebuilding already, this appears to be along the same track. Didn't work the last time so, (doing it but,) doubtful.

 

I am discovering that 10.6.2 "works" but it is a bit buggy. Here I am assuming it is me who hasn't yet realized what is missing. Here are the problems I am having now:

Spotlight: It is acting more like a memory leak of some kind. I have been doing a lot of work with graphics and rendering web pages today in RapidWeaver. I Can work for about 45 minutes to an hour but at a point it starts to stop responding. If it isn't that app, it is another app, and usually I Notice the not responding because I Can get the spotlight window open and type maybe two letters, then it just sits there, spinning the ball and being frozen.

 

Forcing quit, isn't even forcing quit consistently. I Can force quit an app and it still appears in the force quit dialog, when it does quit. SOme of the time apps will force quit but some of the time, they wont.

 

Shutting down: Under this boot configuration it wont shut down all the way. I have to force a hard restart. When I do that I have to boot single user and run APplejack which usually fixes invalid directory count or some such. In truth I don't always run AJ, it will still reboot. When I run AJ it does seem to help in the short term.

 

I am really confused with all the options. THank you for explaining that Vanilla is sufficient for my i7 set up. I thought that all those other versions were more recent and therefore, better. If I am understanding you all I really need then is a plain vanilla installation.

 

I sure wish I was completely stable again. I was hoping to tweak more deeply and start learning about the DSDT patches for my specific system. First things first though.

 

Ongoing thanks for all the help.

Link to comment
Share on other sites

Yeah I got this bug too but its easily fixable. Just copy the background.png file from one of the other themes and u will start with the GUI enabled. My system has been acting so so weird after testing the new script. I thought it was because of the 10.6.5 update but I am not sure as prior to testing the script, all was well. I think the script is full proof, does what its supposed to do. If anyone has some pointers or any weird behavior, please let us know. Thia is really eating me up :angry2: . I hardly shut down my box, I let it sleep and pick up from where I left off. Now when I let it sleep, it just restarts a few minutes after waking up. I have done everything I know...I will test a new drive later and confirm. Cheers Charles

 

CruiSar, thanks for the info, in fact i did that, but wanted to notify Digital_Dreamer about it, so the script gets corrected to work fine out of the box.

 

I reinstalled my system cause i updated to 10.6.5 without removing sleep enabler and all the USB ports where not recogized, used the new script to install in a RAID Volume without problem and updated to 10.6.5 without problem. I'm using the latest sleep enabler and i dinn't noticed those weird restarts that you describe.

 

The only thing that i see is at the startup, where i can read a message that says:

 

Refusing new kext com.apple.iokit.IONetworkFamily. v1.10: already have loaded v.1.9.

Refusing new kext com.apple.iokit.IONetworkFamily. v1.10: already have loaded v.1.9.

 

 

FakeSMC: key info not found MSDS, lenght -6

 

What i don't understand is the utility of the new /extra load method described my Digital_Dreamer, also i don't know why he added so many bootloaders.

Link to comment
Share on other sites

Charles,

Do the reboots happen after a period of inactivity?

I saw your logs, but not sure if anything was showing up right prior to the reboots. Did we see all relevant info?

 

I suppose the "autologout" key was nonexistent in your .GlobalPreferences.plist?

 

kind regards,

MAJ

Hi Maj,

 

The reboots happen when the box is active or not. The auto logout key was non existent. I even deleted all the files in the system configuration folder and left only the bootplist. Nothing seems to work.

 

Yes you are right, nothing shows up in the logs so it makes it really hard to pin point the problem. I even opened it and re-seated the ram and graphics cards, even removed the sata cables and pluged them back in...same old behaviour. I then formatted my test drive and reinstalled and put it to sleep, still at 10.6. I will report back when I wake up in the morning and attempt a wake.

 

CruiSar, thanks for the info, in fact i did that, but wanted to notify Digital_Dreamer about it, so the script gets corrected to work fine out of the box.

 

I reinstalled my system cause i updated to 10.6.5 without removing sleep enabler and all the USB ports where not recogized, used the new script to install in a RAID Volume without problem and updated to 10.6.5 without problem. I'm using the latest sleep enabler and i dinn't noticed those weird restarts that you describe.

 

The only thing that i see is at the startup, where i can read a message that says:

 

Refusing new kext com.apple.iokit.IONetworkFamily. v1.10: already have loaded v.1.9.

Refusing new kext com.apple.iokit.IONetworkFamily. v1.10: already have loaded v.1.9.

 

 

FakeSMC: key info not found MSDS, lenght -6

 

What i don't understand is the utility of the new /extra load method described my Digital_Dreamer, also i don't know why he added so many bootloaders.

I have only 2 kexts running on my system, ionetworkingfamily and fakeSMC, both in extra, everything else is native. Sleep worked fine, in fact I cant even recall the last time I shutdown my box.

 

I also get those 3 errors you posted but the fakeSMC one can be fixed as posted in netkas' blog. The kext by default has debugging mode enabled (true) so you will need to go into the plist file and disable it (false). Those errors wont appear during verbose boot.

 

Cheers

Charles

Link to comment
Share on other sites

 Share

×
×
  • Create New...