This is what I was saying earlier. Updating LS causes a continous boot regardless of what boot.efi is used and regardless of what OS.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Resolved 2006/2007 Mac Pro (1,1/2,1) and OS X Yosemite
- Thread starter Mr. Zarniwoop
- Start date
- Sort by reaction score
This is what I was saying earlier. Updating LS causes a continous boot regardless of what boot.efi is used and regardless of what OS.
I was wondering this too when I tested Pike's boot.efi on my Mavericks install last night. I installed 3.3.4 and it wouldn't reboot. I didn't even get a chance to update. I havebt tried to fix it yet.
Well, I have just updated to LS 3.4.1 in Mavericks (I used to have LS 3.4) and left Tiamo's boot.efi in place. No corruption. No continuous boot. Everything works. I don't understand the situation Peloche and others have reported regarding corruption of Tiamo's boot.efi. Later today I'll try booting Mavericks (with LS 3.4.1 in place, naturally) using Pike's boot.efi and see if I get a continuous boot.
Same trouble here. Sfott is "ready" to rock but I can't boot from the Recovery.
Note as I said that the recovery Base System.dmg IS patched and IS the one I use to asr and create the USB KEY, so the Base System IS OK.
just can't boot at all.
I've md5 BaseSystem in both Recovery HD and the USB and it is not altered by the install process. plist are also "patched" as they should. SO I don't understand yet what the trouble is.
here is what I'll do tonight :
Create another USB same process but without boot.efi (so it will be the ORIGINAL apple boot.efi) but I'll patch everything add the Macmini2,1 for example. - Yes I don't need to, but I want to alter all the files usually altered by the script to be as close as possible from the unsupported situation.
Then I'll use it on a SUPPORTED MacMini.
It should installs Yosemite and create a recovery.
If I can boot Recovery then I should not be far from solution. If I can't boot Recovery, then I'll release a preversion of SFOTT with the Warning of the recovery and will try to figure it out (Am I able to do so? I wonder!).
The Key creation with SFOTT for YOSEMITE works nicelly I've created and installed Yosemite like 3 or 4 times.. since I started...
What I'd like to know is who the Hell did suceed booting the Recovery HD ? (as of course plist, boot.efi and BaseSystem.dmg must be altered)
Note that I also include (replace) sha1 checksum in the OsInstall.mpkg so this should not be the issue.
I'll keep posted and I'll see what I do next.
Any Recovery booted guy please show up
Little Snitch offered me, weeks ago, I think, to update to 3.4.1. I declined at the time, because I couldn't possibly restart my computer at that time. So, I simply accepted the update today. It downloaded the self-installer, installed it and rebooted. I didn't touch boot.efi or anything else. I was prepared for the worst-case scenario (having to resort to Snow Leopard to repair a presumably corrupted boot.efi), but it booted normally. It is working. I'm typing this from my Mac Pro 1,1->2,1 with Mavericks 10.9.5 and Little Snitch 3.4.1 (and still Tiamo's boot.efi). I don't know what to think. Perhaps there's something wrong in your disk. Wrong permissions, perhaps? In a couple of hours I'll try rebooting using Pike's boot.efi (still in Mavericks; I won't install Yosemite until it's officially released).
Last edited:
Maybe the kernel file is also needed in /System/Library/Kernels/ in the BaseSystem.dmg?
Peter you're right, i was testing with the downloaded 3.4.1 instead to accept the update proposal and it works !
When i use the LS proposal to update it doesn't work, so we must know that to update LS we have to download the full dmg file.
It works under Yosemite GM3 and Pike's boot.
OK... So I did it using your guide, and of course it worked fine. But:
1) Why aren't you using the provided CLI tool "createinstallmedia" to make the USB stick?
2) A minor improvement - in the Distribution script, instead of pasting in Machine types, just change the entire "function isSupportedPlatform()" to "return true;"
Using createinstallmedia saves a LOT of steps - it replaces steps 2-17.
I have tested #2 above, and it works.
Well, that's fantastic! So, has this entire problem been caused by a faulty Little Snitch installer that corrupts the boot.efi bootloader? How come they've created two installers, one that works and one that doesn't and, in addition, bricks your computer? Crazy, isn't it? If, as you report, Pike's boot.efi also works (if you don't destroy it with the wrong LS installer), I won't pursue this any further and will instead lazily wait for Yosemite's official release, when I'll follow Hennesie's procedure to install it, and that should be it! Thanks, folks!
Peter,
The update that i've done was on a HDD backup of my SSD (main), so after that i've tried to do the same onto my SSD and ... it didn't work, i've used the exactly same procedure !
???? This is driving me nuts. How can that be? Can anyone else test this either way? As far as I'm concerned, Tiamo's boot.efi doesn't get corrupted in Mavericks after updating to LS 3.4.1. Does Pike's boot.efi get corrupted in Yosemite after updating to LS 3.4.1 on an SSD? Does it get corrupted on an HDD?
With Tiamo's boot.efi on my Mavericks drive and LS 3.3.4 installed I updated to from 3.3.4 to 3.4.1 using the LS Preferences "check for updates". It installed and booted right back up without issue.
This was on a WD Black HDD.
I am going to create a fresh Yosemite install on a HDD using Pike's boot.efi. Then I will install some form of LS <3.4.1 and then update to 3.4.1 and report back.
Ok, so I did a clean install of 10.10 GM1 using my grey version of the boot.efi. Then I installed LS 3.4 and rebooted, which worked fine. Then again I used the "check for updates" button in the LS Preferences and updated to 3.4.1 and it rebooted just like it should. I never touched the boot.efi.
This was on a different HDD than my Mavericks install but it is also a WD Black HDD.
Great. Then, it would appear that this entire episode was caused by some faulty Little Snitch installer that two or three users just happened to come across. All right, folks, all that remains now, I think, is for Yosemite to be officially released and we are set.
Not quite. I am currently running 10.9.5 using the Tiamo boot.efi, (this I combo updated from 10.9.4), and I updated from 3.4 to LS3.4.1 with no problems.
Read my post just before yours. I was only speaking based on what had been posted. I just ran my own test and had no issues at all with either Tiamo's boot.efi and Mavericks or Pike's boot.efi and Yosemite.
I will note that when I installed LS 3.3.4 on the Mavericks HDD it was using Pike's boot.efi and it would not reboot. After placing Tiamo's boot.efi files back on that drive it booted and updated LS fine.
Had an opportunity to definitively test this issue out and my findings concur with yours. Installed a fresh OS X 10.10 Public Beta 4 OS with Pike Alpha Build 6c697848ef on a spare HD. Updated Yosemite to the very latest Public Beta 6. Installed Little Snitch Nighty Build (4201) v3.4 and booted the system with no issue. Installed Little Snitch v3.4.1 overtop/updating v3.4 and again experienced no problems or booting issues.
2006/2007 Mac Pro (1,1/2,1) and OS X Yosemite
Alright then we are good to go, about a day and a half till Yosemite should be released. Although Peloche sill had an issue with updating on a SSD.
Alright then we are good to go, about a day and a half till Yosemite should be released. Although Peloche sill had an issue with updating on a SSD.