• 0

My SATA drives disappears and OS X says "Ejecting error"
1 1

Question

Hi everyone !

 

I have a strange problem. I've just installed Niresh 10.9 on my AMD Athlon 64 x2 2.0 Ghz 2 GB DDR2 RAM. I have 2 SATA drives, one of 80 GB for OS X and another one of 500 GB for Windows 8.1 / Data / Time Machine. Everythings works fine but for an unknown reason, sometimes the second 500 GB drive suddendly disappear from my desktop saying Ejecting Error, try to disconnect the drive by…etc, just as if it was an USB external drive.

 

System Profiler detects it as a SATA drive, the icons aren't yellow, there's just like any normal drive and most of time, I can read/write on them without problems. I did a Time Machine backup without issues.

 

First I thought that it was a Kext conflict between SuperVIAATA.kext and IOAHCIBlockStorageInjector.kext or AHCI_3rdPartySata.kext but I tried to remove the two last of them and be able to boot fine (Removing SuperVIAATA.kext result as "still waiting for root device"). But the problem persist and I can't figure why !

 

The drive works fine on Windows and shows no errors, SMART state is reported as capable and OK and the 80 GB hosting OS X works fine as well.

 

My desktop is a HP dc5750 with the latest BIOS (configure in Native IDE mode - no AHCI available) and beside this very anoying problem under OS X, it works fine.

 

Any ideas ?

 

post-15330-0-57508400-1395508181_thumb.j

Share this post


Link to post
Share on other sites

29 answers to this question

  • 0

Thats a strange Problem Never heard of it.Il do some searching online for it.

1 person likes this

Share this post


Link to post
Share on other sites
  • 0

Thanks, I really hope we can fix it because it make my Hackintosh unusable and it drives me crazy… :blink:

Share this post


Link to post
Share on other sites
  • 0

If it can help, The SATA drive which disappears is a Seagate ST500DM002-1BD142. Another strange thing is that the Seagate disappears but the SATA drive (Western Digital WDC WD800JD-75MSA3)  where OS X is installed remains… for a few minutes until it shows KP due to IOATAFamily IOBlockStorage.

Share this post


Link to post
Share on other sites
  • 0

I would like to know more.

 

1. How many partitions are in dat faulty SATA, with size.

 

2. How could you use time machine backup on a Win OS maintained HDD?

 

3. Are you using any 3rd party NTFS mounter for reading/writing data from NTFS?

 

4. Did you do any naming/labelling to drives in that HDD recently?

 

5. Did you try repair permission (OSX) or chkdks (Win) to that drive jusr before fault occurred ?

 

Last, what is the exact problem are you facing? Your OSX cant mount your SATA HDD?

Share this post


Link to post
Share on other sites
  • 0

Hello !

 

1 SATA 80 GB hosting OS X (1 MBR partition) & 1 SATA 500 GB (1 NTFS partition MBR for Win 8.1 146 GB + 1 NTFS partition MBR for System Reserved 349 MB + 1 NTFS partition MBR for Data 196 GB and 1 HFS+ Mac OS X Extended MBR for Time Machine 158 GB)

 

I use Paragon NTFS 10.0 for Windows to see Mac partitions in Windows but I'm not using Time Machine on Windows, it's just a partition for him. I managed Time Machine backup by booting on the USB Nirech stick and use Time Machine restore.

 

On my first install, I used Tuxera NTFS for managing NTFS partition in OS X but I suspected that it may be the cause of the issue so I disinstalled it but the problem still occurs. When I did a clean fresh re-install of Niresh, I took care not installing it, just to be sure. But still the same.

 

I didn't rename/relabel my drives since the very beginning. I prepared my partitions with an ubuntu-live-CD and I formated them using Disk Utility for OS X & Time Machine and Windows 8.1 installer for NTFS partitions. (i've already did it on a lot of other Hackintosh and it always worked fine)

 

i repaired permissions a lots of times (Full disk and S/L/E and cache during the installation of my own kexts for Graphics and Networking. I ran CHDISKS a countless times in Windows, I even check both HDDs with Seagate and WesternDigital tools for Win and they didn't find any errors. Firmwares are up-to-date and SMART reporting is OK.

 

My exact problem is : I boot normally in OS X, I can see all my drives and partitions, I can read/write to all of them but after a while, all my second SATA drives partitions disappears and OS X says that an error occured while ejecting the disk (just like an USB stick unproperly ejected). My first HDD (hosting OS X) keeps working for a moment and then, OS X freeze. No KP, no nothing, it just freeze.

 

So it's not an hardware issue since Windows has no problems with both drives, it' a software issue and I think there's something wrong with the kexts that manage SATA drives. It appears that it's SuperVIAATA.kext that is mounting the drives and removing it from S/L/E cause a "wait for root device".

 

I can send you my logs and IOReg tree if it can help (too big for an attchment)

 

Thanks :-)

Share this post


Link to post
Share on other sites
  • 0

Well, that is what I was guessing too, never repair permission on MBR HDD from OSX. It not only corrupts the whole FAT also does strange things. One more thing `i would like to add, no matter you format a partition as HFS+ but if windows is installed on this HDD, the whole disk gets MBR technology and causes problem(my own experience, may seniors wont agree). If you have a time machine backup (before this fault occurred will be better) then boot into windows, run chkdsk /f /r on each drive (C drive will need a boot time scan). Let it finish for each drive. Format your OSX 80 GB as HFS+, restore using time machine.

 

It might help.

Share this post


Link to post
Share on other sites
  • 0

What do you mean by "never repair permissions on MBR HDD from OS X" ? You mean never repair permissions of the 80 GB SATA hosting OS X ? Never heard about that ! It implies that everytime I want to install anything (not just kexts) I have to boot Niresh installer to do so ?

The Time Machine partition doesn't need Permissions repairs since it's not hosting an OS X system, just a backup, Disk Utility do not propose to do so anyway.

 

Windows drive doesn't host OS X, just an HFS+ partition and even if it can see it via Paragon HFS+ for Windows, it just ignore it as it would with an Ext4 Linux filesystem. You can't perform chdisk on those partitions.

 

OS X is already on an HFS+ partition, it's just that it's an MBR partition table, not a GUID but it has never been a problem and anyway, it's the only way to use OS X and Windows on a single drive without formatting an existing Windows setup (and that's why Chameleon propose the Boot0hfs option). But again, it's not my case since OS X and Windows are on separete drives. All my others Hackintoshes use a similar setup (OS X on one drive and Win + Time Machine on another one) and it has never been a problem.

 

I really think there's something wrong with the kexts, not the partitions scheme I'm using.

Share this post


Link to post
Share on other sites
  • 0

No I meant dont repair permission on 500gb HDD from OSX. Plz do chkdsk /f /r on each drive.

Share this post


Link to post
Share on other sites
  • 0

But I don't have any OS X install on the 500 GB HD. OS X is hosted on the 80 GB and it's the only one setup I have on this drive. The only HFS+ partition on the 500 GB is the Time Machine backup which does'nt need to be permissions repaired.

 

I think you meant don't Repare disk (and not don't repare permissions) on the 500 GB, right ? And more specificallt, don't try to repare NTFS partitions from OS X, am I correct ?

I never do this, knowing that's it's always better to fix NTFS issues with Windows rather than with OS X.

 

And I did chkdsk /f /r on each NTFS partition of the 500 GB and they're Ok.

Share this post


Link to post
Share on other sites
  • 0

Use tuxera once and test. I had read tuxera FAQ, and they have discussed/addressed this issue. What I will do at your place is to plug in another USB HDD/pen drive and see if it get lost too with SATA. May be fault is only in that HDD.

Share this post


Link to post
Share on other sites
  • 0

Well I disconnected the 500 GB HDD and tryied to pluggin another HDD and it seemes OK so far. So you were right, the 500 GB HDD is someway faulty though I still can't understand why it works fine in Windows and why none of my Hardawre tests I did noticed any problems…

Share this post


Link to post
Share on other sites
  • 0

Oh yes, I did but it was OK with Tuxera too ! It wasn't an "old" HDD", it's a brand new one ;-(

I'm currently formatting it low level and test it with testdisk with ubuntu to see if it's really an harware issue. We'll see, keep in touch ;-)

Share this post


Link to post
Share on other sites
  • 0

Definitely, it was the HDD, lots of bad sectors and same ejection error on another Hackintosh. Bring back tomorrow at the store for replacement.

Share this post


Link to post
Share on other sites
  • 0

I formated my "old" HDD Low Level and there's no more bad sectors (I suspect that it never had any). My Hackintosh seems to run fine so far, i didn't make any HFS+ partitions on the 500 GB HDD, it just holds Windows 8.1 and two others NTFS partitions. I can read/write on them with Tuxera and they does not suddendly disappear like before.

 

I'm facing some system freezes when pluggin USB externel drives for Time Machine Backup but no more ejecting errors so far. So it still a mystery for me about the cause of the ejecting errors.

 

Now, If someone knows how to boot without using cpus=1 boot flag, it would be perfect because it's so frustrating to have a Dual Core capable machine and be forced to use one single core. Tried cpus=2, busratio=10 -force64 dart=0 but while the system boots fine, it starts to act strangely after a while. Grrrrrrrr ;-)

Share this post


Link to post
Share on other sites
  • 0

Finally fix the cpus=1 issue by installing VoodooTSCSyncAMD.kext in E/E & S/L/E. Now I'm enjoying full Dual Core Support ! Time clock is not always accurate and some icons bounces in the dock but nothing really anoying…

 

Thanks for your answers and help guys !

Share this post


Link to post
Share on other sites
  • 0

Well back again folks !

 

Guess what, the error came back and whatever HDD I use, it's still the same…

 

It's driving me crazy ! I suspect that it's USB related because the error occurs (but I'm not sure) when I plug an external USB 3.0 HDD to perform a Time Machine backup. Seems like at a certain point, the OS goes crazy with read/write operations. But I don't understand why it's my internal HDD that's ejected and not my USB one. It's non-sens !

Share this post


Link to post
Share on other sites
  • 0

Do you have any Legacy support or something related to it feature in BIOS?

 

Change its value and try. 

Share this post


Link to post
Share on other sites
  • 0

Hello Aman ! Thanks for your support :-)

 

Unfortunately, I don't have any USB related option in my BIOS (It's a very old and restricted HP BIOS, it can't even let me choose which SATA drive I want to boot from xangry.png.pagespeed.ic.Fjp9rDDtF5.png )…

Share this post


Link to post
Share on other sites
  • 0

HP...hmm I had read about hp boot flag but not sure what its purpose is.

 

Try this flag too.

 

BTW, why do you left tuxera if it was working well?

Share this post


Link to post
Share on other sites
  • 0

Yes I letf Tuxera and it works fine (but with or w/o it, I'va got the same error…)

 

What is the HP boot flag ? Is it just HP in chameleon boot screen ?

Share this post


Link to post
Share on other sites
  • 0

Ok, just added the boot flag to org.chameleon.plist (it's now hp npci=0x2000 and nothing else)… We'll see and I'll do some stress-tests to check if it's useful rolleyes.gif.pagespeed.ce.qeSuMdX8nQ.gif

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.
1 1