MP 1,1-5,1 MacPro 5.1 & high capacity hard drives

[krakman]

Yes this needs to be a Sticky thread with a list of compatible HDD on the first post.

My experience:

Shucked WD 10TB - does not survive a warm reboot - requires new drive sled
Shucked WD 12TB WDC WD120EDBZ - works fine on reboot - requires new drive sled

 

[arw]

Afaik all Helium disks require a new drive sled, so I tried the following in the optical bay just for the sake of this post:

• Working, also on reboot:
  • 18 TB Toshiba (MG09ACA18TE)
• Working (without 3rd pin mod), doesn't survive reboot:
  • 8 TB HGST/WD Ultrastar HE10 (DC HC510 / HUH721008ALE600)
  • 10 TB WD Red (WD100EFAX)
  • 12 TB HGST/WD Ultrastar HE12 (HUH721212ALE600)

 

[amarmot]

I was forced for various reasons to update to APFS, which in turn required replacing the 4 HDDs in the drive bays with SSDs (long story). In any case, I have not had this problem with SSDs.

 

[bigpoppa]

Any idea why a large hard drive slid into an external USB connected box works just fine but when I try to hook the drive up directly through SATA the system can't see it.

 

[h9826790]

Any idea why a large hard drive slid into an external USB connected box works just fine but when I try to hook the drive up directly through SATA the system can't see it.

Most likely due to the HDD pin 3 is the PWDIS pin

 

[amarmot]

Any idea why a large hard drive slid into an external USB connected box works just fine but when I try to hook the drive up directly through SATA the system can't see it.

I remember reading a lot of speculation about this, but I'm not sure there was a clear conclusion. A dominant hypothesis was that big drives may take slightly longer to spin up. The boot routine may time out if it does not immediately detect a drive on the SATA 2 bus, so moves on until it finds a system. Drives connected via native USB 2 are not a problem because this interface is designed to detect when a new device is attached. But SATA 2 is not hot swapable, so perhaps if a drive is not detected in time, it never will be. Cold boot may provide more time before the boot sequence looks for the drives, so less chance of missing it.

 

[amarmot]

Most likely due to the HDD pin 3 is the PWDIS pin

I know that was a problem for many people. In my case the usual solutions (eg taping the pin with kapton) didn't help. PWDIS also doesn't readily explain why the drives were booting fine during a cold restart.

 

[krakman]

I know that was a problem for many people. In my case the usual solutions (eg taping the pin with kapton) didn't help. PWDIS also doesn't readily explain why the drives were booting fine during a cold restart.

I had this problem with a 10TB WD drive. I taped up pin 3 to get it working in the Mac Pro. it would mount on cold boot but would not survive a warm reboot.

Because it was just used to archive data and not as a boot drive I used a app (jettison or mountain) to "prevent mount on startup". This allowed the drive to survive the warm reboot and I could mount it on the desktop whenever I needed it.

After a few years ithe drive became noisy and started clicking so I replaced it with a 12TB WD which works perfectly without any need for any modifications or software.

 

[czs]

General question to the community:

In terms of internal drives, does anyone know what the cMP5,1 does differently between a power-on start and a warm-reboot? To my way of thinking, the answer to the not-seen-warm-reboot part of this lies somewhere within that delta given the the drives do work fine after a power-on boot.

 

[arw]

General question to the community:

In terms of internal drives, does anyone know what the cMP5,1 does differently between a power-on start and a warm-reboot? To my way of thinking, the answer to the not-seen-warm-reboot part of this lies somewhere within that delta given the the drives do work fine after a power-on boot.

According to the link in post #43:

It now looks like this is a drive-availability issue. Earlier OS X versions mount these drives fine on reboot, while later macOS versions do not. It may be related to ability to boot from a RAID array. On reboot, later macOS versions expect drives to be available right away. If they aren't ready fast enough, they're ignored. A cold boot allows more time to spin up (due to POST?) avoiding the glitch. Apple could definitely fix it, as older OS versions are unaffected. HGST / WD might fix it, by reporting 'ready' sooner, even if the drive hasn't finished spinning up.

But it's above my understanding how that could be the case as my disks don't spin down and I even hear/feel the reading head move before the OC boot picker is displayed. And yet they are invisible to the Mac.
Jettison didn't make a difference. Haven't tried Mounty.

 

[DigitalSurgery]

Just installed 2x 16TB Western Digital Ultrastar DC HC550 (SATA 6Gb/s) WUH721816ALE6L0 512e ISE in a JBOD softraid with Raid Assistant. Upgrading from 4x 6TB HGST which worked fine for years.

All works on a cold boot but not on a warm one. Used gaffer tape for mounting with the traditional sleds. Placed the redundant screws also with piece of gaffer behind the handle. All works fine on Monterey 12.5.1 😎

Cheers!

 

[dickiedunn]

Just installed 2x 16TB Western Digital Ultrastar DC HC550 (SATA 6Gb/s) WUH721816ALE6L0 512e ISE in a JBOD softraid with Raid Assistant. Upgrading from 4x 6TB HGST which worked fine for years.

All works on a cold boot but not on a warm one. Used gaffer tape for mounting with the traditional sleds. Placed the redundant screws also with piece of gaffer behind the handle. All works fine on Monterey 12.5.1 😎

Cheers!

View attachment 2078714
View attachment 2078715
View attachment 2078718
View attachment 2078722
View attachment 2078723

 

[dickiedunn]

Thanks, I like the gaffer tape idea with the traditional threads. Any long term experience with how that will hold up over time? Or I guess we'll find out. I'm not sure what you mean by them not working on a warm boot though.

 

[Soba]

Just installed 2x 16TB Western Digital Ultrastar DC HC550 (SATA 6Gb/s) WUH721816ALE6L0 512e ISE in a JBOD softraid with Raid Assistant. Upgrading from 4x 6TB HGST which worked fine for years.

All works on a cold boot but not on a warm one. Used gaffer tape for mounting with the traditional sleds. Placed the redundant screws also with piece of gaffer behind the handle. All works fine on Monterey 12.5.1 😎

Cheers!

I do not recommend this approach.

Tape adhesive does not hold up long term, especially in situations where it's exposed to heat or oscillating temperature changes. Over time, that tape is likely to fail and the drive will fall loose from the sled at an angle, which will torque the drive connector on your backplane/main board and destroy the board, your drive, your data, and possibly other components in your system in the process.

If you're going to use a poor man's solution like this, try wrapping the drive tightly in a complete circle of tape (more than once) so the length of tape acts as a restraint (or safety net, if you will) against the drive falling. Better, but still not a great solution.

A great solution: Buy a new tray. They cost less than $20 and that's a tiny investment to guarantee the safety of your (very expensive) equipment and your data!

 

[zachek]

Thanks, I like the gaffer tape idea with the traditional threads. Any long term experience with how that will hold up over time? Or I guess we'll find out. I'm not sure what you mean by them not working on a warm boot though.

He means it won't recognize the drives after a restart (warm boot), but if the machine is off and booted (cold boot) it *will* recognize them normally.

 

[DigitalSurgery]

I do not recommend this approach.

Tape adhesive does not hold up long term, especially in situations where it's exposed to heat or oscillating temperature changes. Over time, that tape is likely to fail and the drive will fall loose from the sled at an angle, which will torque the drive connector on your backplane/main board and destroy the board, your drive, your data, and possibly other components in your system in the process.

If you're going to use a poor man's solution like this, try wrapping the drive tightly in a complete circle of tape (more than once) so the length of tape acts as a restraint (or safety net, if you will) against the drive falling. Better, but still not a great solution.

A great solution: Buy a new tray. They cost less than $20 and that's a tiny investment to guarantee the safety of your (very expensive) equipment and your data!

I do not recommend this approach.

Tape adhesive does not hold up long term, especially in situations where it's exposed to heat or oscillating temperature changes. Over time, that tape is likely to fail and the drive will fall loose from the sled at an angle, which will torque the drive connector on your backplane/main board and destroy the board, your drive, your data, and possibly other components in your system in the process.

If you're going to use a poor man's solution like this, try wrapping the drive tightly in a complete circle of tape (more than once) so the length of tape acts as a restraint (or safety net, if you will) against the drive falling. Better, but still not a great solution.

A great solution: Buy a new tray. They cost less than $20 and that's a tiny investment to guarantee the safety of your (very expensive) equipment and your data!

Hi Soba, I know from experience that proper gaffers tape will hold, even if it will get hot. But you have to use real Gaffa tape, preferably from a brand like Nichiban not a fake one like duct tape. The difference is described here:

Gaffers Tape vs. Duct Tape

<p>Have you ever wondered why gaffers tape is more expensive than duct tape?

www.findtape.com

Of course a dedicated tray is a better option, but this solution works fine for me atm. Cheers!

 

[DigitalSurgery]

He means it won't recognize the drives after a restart (warm boot), but if the machine is off and booted (cold boot) it *will* recognize them normally.

Exactly, and probably when not combined in a soft raid they will show up in a warm boot. I’ve heard that it has to do with the drives/raid not showing up fast enough when warm booting to be recognized. My other 10TB drive is normally recognized on a warm boot.

 

[macguru9999]

Some sata 3 drives are not backwards compatible with sata 2 in the 5,1 trays, those will not show up on warm boot. like the samsung qvo ssd drives etc. Also its crazy having heavy drives taped in like that, get the correct trays from OWC. I would move the incompatible drives to usb 3 housings and replace them with drives that work properly in 5,1s. Drives that others have verified work. seagate 8tb spinners work but i suppose fewer and fewer new models will as they expect to be used on sata 3 buses.

 

[DigitalSurgery]

Some sata 3 drives are not backwards compatible with sata 2 in the 5,1 trays, those will not show up on warm boot. like the samsung qvo ssd drives etc. Also its crazy having heavy drives taped in like that, get the correct trays from OWC. I would move the incompatible drives to usb 3 housings and replace them with drives that work properly in 5,1s. Drives that others have verified work. seagate 8tb spinners work but i suppose fewer and fewer new models will as they expect to be used on sata 3 buses.

Thanks for responding. Yeah I know it’s looks ‘crazy’ to use Gaffa tape, but you would be surprised how often tape is being used in high-end applications nowadays. And I don’t want to spend money just for cosmetical reasons while this works perfectly. Remember, the drives are still in place by two screws, the tape is just to hold the drives ‘up’ and it does the trick great. Personally being a fan of DIY solutions I prefer to stick 😉 with this method, but anyone is free to do otherwise.

 

[zedex]

General question to the community:

In terms of internal drives, does anyone know what the cMP5,1 does differently between a power-on start and a warm-reboot? To my way of thinking, the answer to the not-seen-warm-reboot part of this lies somewhere within that delta given the the drives do work fine after a power-on boot.

I too am all for certainty and specificity and so without further ado:

FROM PAGE 32 OF THE APPLE TECHNICIANS MANUAL..

"When the computer is started up after being fully shut down, a self test in the computer’s ROM is automatically run. (The test is not run if the computer is only restarted.)"

The fact is, I always have to go to Alex if I need an answer that is 'precisely correct'.

I think "POST" stands for Pre-Operating-System-Test (or maybe Pre-Os-Self-Test). Either way you finally have a 'precisely correct' answer to the question posed above..

COLD BOOT runs "POST"
WARM BOOT does *not* run "POST"

Amazing huh! I tripped across this paragraph earlier this morning so I am happy I could help

 

[m1maverick]

I think "POST" stands for Pre-Operating-System-Test (or maybe Pre-Os-Self-Test). Either way you finally have a 'precisely correct' answer to the question posed above..

COLD BOOT runs "POST"
WARM BOOT does *not* run "POST"

Amazing huh! I tripped across this paragraph earlier this morning so I am happy I could help

POST = Power On Self Test

 

[czs]

COLD BOOT runs "POST"
WARM BOOT does *not* run "POST"

Thanks - makes senses so if that is truly the only delta, then the questions now are: (a) can it be fixed; or (b) is there any way to have the POST run even know there is no power-on event? I believe both are doubtful - "A" because I believe POSTs reside in true-ROM and "B" because I would bet that code is hardwired to only get accessed during an actual power-on event with no other way to trigger its execution...

 

[MarkC426]

Short of going through the 300+ posts here, is there a 'rule of thumb' as to which capacity drives have the holes in different positions. I am looking at Seagate Baracuda 4TB/6TB/8TB HDD, and want the largest without a new bracket.

 

[tsialex]

Short of going through the 300+ posts here, is there a 'rule of thumb' as to which capacity drives have the holes in different positions. I am looking at Seagate Baracuda 4TB/6TB/8TB HDD, and want the largest without a new bracket.

Both models of Seagate 5TB and 6TB that I shucked from external USB drives have the standard screw arrangement and work correctly.

5TB part number is ST5000DM000.

 

[arw]

Short of going through the 300+ posts here, is there a 'rule of thumb' as to which capacity drives have the holes in different positions. I am looking at Seagate Baracuda 4TB/6TB/8TB HDD, and want the largest without a new bracket.

I remember reading due to the additional sealing requirements all helium drives moved said holes away from the middle.
That doesn‘t mean there aren‘t also non-Helium drives with the new holes (to fit slightly larger or more platters) but Helium should be a dead giveaway.

edit: seems like 4 disks is the limit for Seagate:
- The 3 disk 6TB ST6000DM003 and the 4 disk 8TB ST8000DM004 have the standard mounting points

Seagate BARRACUDA ST8000DM004 Product Manual (Page 23 of 34) | ManualsLib

www.manualslib.com

- Both 6 disk 6TB ST6000AS0002 and 6 disk 8TB ST8000AS0002 have the new ones

Drive Mounting - Seagate ST8000AS0002 Product Manual [Page 23]

Seagate ST8000AS0002 Manual Online: Drive Mounting. Users can mount the drive in any orientation using four screws in the side-mounting holes or four screws in the bottom-mounting holes. Refer to for drive mounting dimensions. Follow these important mounting Figure 3 precautions when...

www.manualslib.com