2015 MBP freezes under 10.11.6

[Robert Peirce]

Apparently this is a widely known problem. Unfortunately, I didn't know
about it before buying one for my wife last month.

She has the latest 2015 15" MBP with the SSD drive running 10.11.6. [I
have a similar 2012 model that has no problems.] Google indicates the
problem cropped up with 10.11.4 and nobody really knows what to do about
it. However, I have determined that the problem doesn't exist with Safe
Boot and I am wondering what a regular boot starts up that Safe Boot
doesn't.

The problem only shows up on her computer when awakening from sleep.
With this problem others have encountered the fan running; we only saw
that once (so far). Others have had problems relating to specific
programs but that is not the case here. Under regular boot we tried
killing any running programs. Then we added programs that start on
login. Finally we just logged off to be sure whatever was happening was
at the system level. In every case it froze on wakening. The cursor
arrow would eventually appear but it didn't do anything.

Safe Boot is different. We have not encountered any problems no matter
what we have done. Therefore, we got to wondering what Safe Boot
DOESN'T start. Activity Monitor was no help. There are too many things
running, even under Safe Boot. We thought somebody might know what
doesn't run and that might lead to the source of the problem.

[Davoud]

Post by Robert Peirce
Apparently this is a widely known problem. Unfortunately, I didn't know
about it before buying one for my wife last month.
She has the latest 2015 15" MBP with the SSD drive running 10.11.6

Any thought of upgrading to the current version of Mac OS? If it
doesn't fix the problem, which it might well do, it probably won't make
it worse.

[Robert Peirce]

Post by Davoud
Any thought of upgrading to the current version of Mac OS? If it
doesn't fix the problem, which it might well do, it probably won't make
it worse.

I am in VA, waiting until I return to PA. I think it should be okay. I
haven't heard of anything terribly disturbing.

I have narrowed the problem down to the display going to sleep. If you
set the computer up to stay awake when the display sleeps it still
freezes. If you never let the display sleep it doesn't.

I was expecting "infant mortality" but the machine passed the Apple
hardware test, which may not be extensive enough. Anyhow, I thought I
would check at the local Apple store on return to see if they know anything.

[Jolly Roger]

Post by Robert Peirce
Apparently this is a widely known problem. Unfortunately, I didn't know
about it before buying one for my wife last month.
She has the latest 2015 15" MBP with the SSD drive running 10.11.6. [I
have a similar 2012 model that has no problems.] Google indicates the
problem cropped up with 10.11.4 and nobody really knows what to do about
it. However, I have determined that the problem doesn't exist with Safe
Boot and I am wondering what a regular boot starts up that Safe Boot
doesn't.
The problem only shows up on her computer when awakening from sleep.
With this problem others have encountered the fan running; we only saw
that once (so far). Others have had problems relating to specific
programs but that is not the case here. Under regular boot we tried
killing any running programs. Then we added programs that start on
login. Finally we just logged off to be sure whatever was happening was
at the system level. In every case it froze on wakening. The cursor
arrow would eventually appear but it didn't do anything.
Safe Boot is different. We have not encountered any problems no matter
what we have done. Therefore, we got to wondering what Safe Boot
DOESN'T start. Activity Monitor was no help. There are too many things
running, even under Safe Boot. We thought somebody might know what
doesn't run and that might lead to the source of the problem.

Could be a malfunctioning third-party kernel extension. What do you have in
/Library/Extensions?

[Robert Peirce]

Post by Jolly Roger
Could be a malfunctioning third-party kernel extension. What do you have in
/Library/Extensions?

There are 12 extensions but they all date to 2015 or earlier. These
were migrated from the computer we replaced. That computer and my own
2012 model worked/works with these extensions.

I'm pretty sure it has something to do with the display sleeping but I
don't know what.

[Jolly Roger]

Post by Robert Peirce

Post by Jolly Roger
Could be a malfunctioning third-party kernel extension. What do you
have in /Library/Extensions?

There are 12 extensions but they all date to 2015 or earlier.

Okay, but I asked for a list of them. Are they a secret for some reason?

Post by Robert Peirce
These were migrated from the computer we replaced.

Good to know. It's possible one or more were not migrated completely or
is malfunctioning after being migrated for some reason.

Post by Robert Peirce
That computer and my own 2012 model worked/works with these
extensions.

But that's another computer where they were installed with their native
installer rather than being migrated.

Post by Robert Peirce
I'm pretty sure it has something to do with the display sleeping but I
don't know what.

Kernel extensions are notorious for misbehaving during sleep/wake
events. That's why I asked in the first place. ; )

So about that list...

[Robert Peirce]

Post by Jolly Roger

Post by Robert Peirce
I'm pretty sure it has something to do with the display sleeping but I
don't know what.

Kernel extensions are notorious for misbehaving during sleep/wake
events. That's why I asked in the first place. ; )
So about that list...

I misunderstood. Here's the list, with comparisons:

On problem MBP On Mini On my MBP
AC3x.kext ACS6x.kext ACS6x.kext
ArcMSR.kext
ATTOCelerityFC8.kext
ATTOExpfessSASHBA2.kext
ATTOExpressSASRAID2.kext
CalDigitHPProDrv.kext
DroboTBT.kext DroboTBT.kext
EPSONUSBPrintClass.kext
HighPointOP.kext
HighPointRR.kext
hp_io_enabler_compound.kext
PromiseSTEX.kext
SoftRAID.kext
TCNear.kext Same
TrustedDataSCSIDriver.kext Same

All have the same name, except as noted. I didn't byte compare these
but ACS3x.kext, whatever that is, is unique to the problem computer.
The others have ACS6x.kext. I have no idea if that matters.

[David Empson]

Post by Jolly Roger

Post by Robert Peirce
I'm pretty sure it has something to do with the display sleeping but I
don't know what.

Kernel extensions are notorious for misbehaving during sleep/wake
events. That's why I asked in the first place. ; )
So about that list...

On problem MBP On Mini On my MBP
AC3x.kext ACS6x.kext ACS6x.kext
ArcMSR.kext
ATTOCelerityFC8.kext
ATTOExpfessSASHBA2.kext
ATTOExpressSASRAID2.kext
CalDigitHPProDrv.kext
DroboTBT.kext DroboTBT.kext
EPSONUSBPrintClass.kext
HighPointOP.kext
HighPointRR.kext
hp_io_enabler_compound.kext
PromiseSTEX.kext
SoftRAID.kext
TCNear.kext Same
TrustedDataSCSIDriver.kext Same
All have the same name, except as noted. I didn't byte compare these
but ACS3x.kext, whatever that is, is unique to the problem computer.
The others have ACS6x.kext. I have no idea if that matters.

I also have an "ACS6x.kext" on two Macs, one running 10.11.6 and the
other running 10.12.1. Get Info says mine is version 3.1.8, dated June
2014, and its copyright message says "Copyright (c) 2004-2014 Accusys,
Inc."

I haven't installed any third party software which might include drivers
on the Mac running 10.11.6, so it looks it is installed by Apple as part
of the standard system.

Confirmed. It is in the installers for macOS Sierra 10.12.1, OS X El
Capitan 10.11.6 and OS X Yosemite 10.10.5. It isn't in the installer for
OS X Mavericks 10.9.5.

I didn't see AC3x.kext in the installer for any of those OS versions.

[Jolly Roger]

Post by David Empson
I didn't see AC3x.kext in the installer for any of those OS versions.

That was likely a typo (ACS3x.kext).

[David Empson]

Post by Jolly Roger

Post by David Empson
I didn't see AC3x.kext in the installer for any of those OS versions.

That was likely a typo (ACS3x.kext).

I didn't find anything with a similar name in /Library/Extensions in the
installers: nothing else with a name starting with AC apart from
ACS6x.kext in Yosemite and later.

In any case, the absence of ACS6x.kext on the misbehaving MBPro is odd,
because it is part of the OS install.

[Jolly Roger]

Post by David Empson

Post by Jolly Roger

Post by David Empson
I didn't see AC3x.kext in the installer for any of those OS
versions.

That was likely a typo (ACS3x.kext).

I didn't find anything with a similar name in /Library/Extensions in
the installers: nothing else with a name starting with AC apart from
ACS6x.kext in Yosemite and later.

It could be named ACS3x.kext for the 32-bit version and ACS64x.kext for
the 64-bit version.

Post by David Empson
In any case, the absence of ACS6x.kext on the misbehaving MBPro is
odd, because it is part of the OS install.

Considering that this system was migrated from another Mac (unmentioned
in the initial post), it's possible whatever version got migrated is not
compatible with this newer computer. An incompatible kernel extension
could definitely cause sleep/wake malfunctions.

[Jolly Roger]

Post by Jolly Roger

Post by David Empson

Post by Jolly Roger

Post by David Empson
I didn't see AC3x.kext in the installer for any of those OS
versions.

That was likely a typo (ACS3x.kext).

I didn't find anything with a similar name in /Library/Extensions in
the installers: nothing else with a name starting with AC apart from
ACS6x.kext in Yosemite and later.

It could be named ACS3x.kext

Oops, brain fart. I meant ACS32x.kext.

[Jolly Roger]

Post by Robert Peirce

Post by Jolly Roger

Post by Robert Peirce
I'm pretty sure it has something to do with the display sleeping but I
don't know what.

Kernel extensions are notorious for misbehaving during sleep/wake
events. That's why I asked in the first place. ; )
So about that list...

On problem MBP On Mini On my MBP
AC3x.kext ACS6x.kext ACS6x.kext
ArcMSR.kext
ATTOCelerityFC8.kext
ATTOExpfessSASHBA2.kext
ATTOExpressSASRAID2.kext
CalDigitHPProDrv.kext
DroboTBT.kext DroboTBT.kext
EPSONUSBPrintClass.kext
HighPointOP.kext
HighPointRR.kext
hp_io_enabler_compound.kext
PromiseSTEX.kext
SoftRAID.kext
TCNear.kext Same
TrustedDataSCSIDriver.kext Same
All have the same name, except as noted. I didn't byte compare these
but ACS3x.kext, whatever that is, is unique to the problem computer.
The others have ACS6x.kext. I have no idea if that matters.

hp_io_enabler_compound.kext seems to be part of HP 8600e printer
software. It's been known to cause issues:

<https://discussions.apple.com/thread/7286706?start=0&tstart=0>

Are you actually using an HP printer with the problem MBP? Was
this migrated over? You may want to try downloading an uninstaller from
HP for that software package to remove all vestiges of it from the
machine in case it's the cause of the problem.

[Robert Peirce]

Post by Jolly Roger
Are you actually using an HP printer with the problem MBP? Was
this migrated over? You may want to try downloading an uninstaller from
HP for that software package to remove all vestiges of it from the
machine in case it's the cause of the problem.

I have an HP Office Jet 6000 wireless printer that all my computers use
in PA. In VA I have an USB Epson on my MBP that I share with the other MBP.

However, hp_io_enabler_compound.kext is identical on both machines.

[Robert Peirce]

On problem MBP On Mini On my MBP
AC3x.kext ACS6x.kext ACS6x.kext

This is wrong. I just checked and it now says AC6x. I may have
miss-typed it but now I don't know.

All have the same name, except as noted. I didn't byte compare these
but ACS3x.kext, whatever that is, is unique to the problem computer. The
others have ACS6x.kext. I have no idea if that matters.

I ran diff -r on all the directories between the problem MBP and my MBP.
They are all the same.

I'm not sure to what degree diff matches cmp. I suppose the bytes must
be different for the line to be different. I used to have a directory
compare utility that used cmp but I lost it years ago.

[Robert Peirce]

Post by Robert Peirce

On problem MBP On Mini On my MBP
AC3x.kext ACS6x.kext ACS6x.kext

This is wrong. I just checked and it now says AC6x. I may have
miss-typed it but now I don't know.

Not being sure if that ACS3x was correct or not, I have allowed the
problem MBP to go to sleep to see if it freezes. However, as an old
friend used to say, "You can never prove the absence of a bug." It may
wake up but still freeze later.

[Robert Peirce]

Post by Robert Peirce

Post by Robert Peirce

On problem MBP On Mini On my MBP
AC3x.kext ACS6x.kext ACS6x.kext

This is wrong. I just checked and it now says AC6x. I may have
miss-typed it but now I don't know.

Not being sure if that ACS3x was correct or not, I have allowed the
problem MBP to go to sleep to see if it freezes. However, as an old
friend used to say, "You can never prove the absence of a bug." It may
wake up but still freeze later.

I tried this several times and it still froze. Since all the kernel
extensions match all three computers it must be something else.

[Jolly Roger]

Post by Robert Peirce

Post by Robert Peirce

Post by Robert Peirce

On problem MBP On Mini On my MBP
AC3x.kext ACS6x.kext ACS6x.kext

This is wrong. I just checked and it now says AC6x. I may have
miss-typed it but now I don't know.

Not being sure if that ACS3x was correct or not, I have allowed the
problem MBP to go to sleep to see if it freezes. However, as an old
friend used to say, "You can never prove the absence of a bug." It may
wake up but still freeze later.

I tried this several times and it still froze. Since all the kernel
extensions match all three computers it must be something else.

I forget: Have you made note of the time of the freeze, then looked at the
system log for any error messages around that time?

[Robert Peirce]

Post by Jolly Roger
I forget: Have you made note of the time of the freeze, then looked at the
system log for any error messages around that time?

No, I have not. I have great difficulty working with logs. I can never
figure out which one to examine and I really don't understand what I am
looking at/for.

[Jolly Roger]

Post by Robert Peirce

Post by Jolly Roger
I forget: Have you made note of the time of the freeze, then looked at the
system log for any error messages around that time?

No, I have not. I have great difficulty working with logs. I can never
figure out which one to examine

The system log is at /var/log/system.log.

Post by Robert Peirce
and I really don't understand what I am
looking at/for.

Messages with a time stamp at or near the time of the problem. See above.

[Robert Peirce]

I updated the computer to Sierra and that seems to have solved the
problem. Of course, one can never know that for sure but so far so good.

[Robert Peirce]

After updating to Sierra my wife's 2015 MBP stopped freezing but now my
Mid-2012 MBP freezes! I knew about when I logged off and when I
rebooted so I looked at the console log. There were too many messages
to mention and most of them either I didn't understand or didn't have
any idea what to do about. For example:

Nov 23 11:19:56 robert-peirces-computer com.apple.xpc.launchd[1]
(com.apple.VoiceOver.running): The HideUntilCheckIn property is an
architectural performance issue. Please transition away from it.

OK, but how?

The following occurred between the time I logged off and tried to log
back on and had to reboot. I'm thinking the 11:22 message was the last
one before it froze. 11:32 may have been when I started trying to logon
and 11:40 seems to be when I rebooted.

Nov 23 11:22:49 robert-peirces-computer com.apple.xpc.launchd[1]
(com.apple.security.idskeychainsyncingproxy[9540]): Service did not exit
5 seconds after SIGTERM. Sending SIGKILL.
Nov 23 11:32:38 robert-peirces-computer syslogd[72]: ASL Sender Statistics
Nov 23 11:40:42 localhost bootlog[0]: BOOT_TIME 1479919242 0

[Jolly Roger]

Post by Robert Peirce
After updating to Sierra my wife's 2015 MBP stopped freezing but now my
Mid-2012 MBP freezes! I knew about when I logged off and when I
rebooted so I looked at the console log. There were too many messages
to mention and most of them either I didn't understand or didn't have
Nov 23 11:19:56 robert-peirces-computer com.apple.xpc.launchd[1]
(com.apple.VoiceOver.running): The HideUntilCheckIn property is an
architectural performance issue. Please transition away from it.
OK, but how?
The following occurred between the time I logged off and tried to log
back on and had to reboot. I'm thinking the 11:22 message was the last
one before it froze. 11:32 may have been when I started trying to logon
and 11:40 seems to be when I rebooted.
Nov 23 11:22:49 robert-peirces-computer com.apple.xpc.launchd[1]
(com.apple.security.idskeychainsyncingproxy[9540]): Service did not exit
5 seconds after SIGTERM. Sending SIGKILL.
Nov 23 11:32:38 robert-peirces-computer syslogd[72]: ASL Sender Statistics
Nov 23 11:40:42 localhost bootlog[0]: BOOT_TIME 1479919242 0

You haven't described the nature of the freeze. When does it happen?
What are you doing with the computer at the time? Does the cursor move
when you move the mouse/trackpad? Etc.

[Robert Peirce]

Post by Jolly Roger
You haven't described the nature of the freeze. When does it happen?
What are you doing with the computer at the time? Does the cursor move
when you move the mouse/trackpad? Etc.

It is the same problem as my wife's had. If you let the machine go to
sleep it may not wake up. Usually this happens if I log off for the
night. This morning I just left it for a couple of hours and when I
came back all I had was the screen saver. I couldn't even get the mouse
icon to wake up.

My wife's computer seemed to freeze every time you let it sleep under
10.11.6. Mine only does it sometimes. I have not kept track of what
percentage of the time it freezes but it is less than 100%. Her's
always froze.

[Jolly Roger]

Post by Robert Peirce

Post by Jolly Roger
You haven't described the nature of the freeze. When does it happen?
What are you doing with the computer at the time? Does the cursor move
when you move the mouse/trackpad? Etc.

It is the same problem as my wife's had. If you let the machine go to
sleep it may not wake up. Usually this happens if I log off for the
night. This morning I just left it for a couple of hours and when I
came back all I had was the screen saver.

Did the screen saver continue animations after you tried to wake up the
computer? Or did it stop and display something else? Or was the screen
static (frozen)?

Post by Robert Peirce
I couldn't even get the mouse icon to wake up.

Are you trying to say that the cursor wouldn't move when you moved the
mouse?

Did you try pressing the space bar or another key on the keyboard?

Did the computer respond in *any* way to your efforts to wake it up?

Post by Robert Peirce
My wife's computer seemed to freeze every time you let it sleep under
10.11.6. Mine only does it sometimes. I have not kept track of what
percentage of the time it freezes but it is less than 100%. Her's
always froze.

There's got to be something in common between the two of them. Often
kernel panics are caused by misbehaving drivers. So perhaps they both
have the same third-party kernel extension installed, or perhaps both
are connected to the same make/model external USB device(s that are
malfunctioning during sleep/wake events). If the kernel is panicking
there should be a corresponding panic log with details about the panic;
IIRC you posted one of them for the other machine...

[Robert Peirce]

Post by Jolly Roger
Did the screen saver continue animations after you tried to wake up the
computer? Or did it stop and display something else? Or was the screen
static (frozen)?

Frozen

Post by Jolly Roger

Post by Robert Peirce
I couldn't even get the mouse icon to wake up.

Are you trying to say that the cursor wouldn't move when you moved the
mouse?
Did you try pressing the space bar or another key on the keyboard?
Did the computer respond in *any* way to your efforts to wake it up?

The cursor didn't even appear. I pressed several different keys with no
result. Eventually I had to power down and back up

Post by Jolly Roger
There's got to be something in common between the two of them. Often
kernel panics are caused by misbehaving drivers. So perhaps they both
have the same third-party kernel extension installed, or perhaps both
are connected to the same make/model external USB device(s that are
malfunctioning during sleep/wake events). If the kernel is panicking
there should be a corresponding panic log with details about the panic;
IIRC you posted one of them for the other machine...

I have some add-on stuff but her machine is about as clean as can be.
The only thing we have in commons is we both use the Microsoft Wireless
Mobile Mouse 3000, but disconnecting it doesn't seem to help. The
interesting thing is I had not problem with 10.11 and she die. Now she
has no problem with 10.12 but I do. However, it isn't every time for
me. I can't even tell in advance if it will freeze. Sometime it does
and sometimes it doesn't. When I looked at the console logs I saw
nothing that I cold identify as any kind of a panic.

[Robert Peirce]

I have an update based on more experience. As near as I can tell the
ONLY time this happens is if I log off and let the screen go to sleep
[black]. There may be a time period required but I have not determined
that. If I don't log off and the screen goes to sleep it seems always
to wake up, even overnight.

When I attempt to wake the computer to log in the screen stays black.
However, I have a Kensington USB numeric pad that has indicator lights
and they stay on. Therefore, as near as I can tell the computer is
still functioning in some way. The screen just won't wake up.

I wondered why this doesn't happen with my wife's 2015 MBP but
remembered she shuts her computer off when she is not using it. We're
going to try logging off and leaving it on at night to see what happens.

[Jolly Roger]

Post by Robert Peirce
I have an update based on more experience. As near as I can tell the
ONLY time this happens is if I log off and let the screen go to sleep
[black]. There may be a time period required but I have not determined
that. If I don't log off and the screen goes to sleep it seems always
to wake up, even overnight.
When I attempt to wake the computer to log in the screen stays black.
However, I have a Kensington USB numeric pad that has indicator lights
and they stay on. Therefore, as near as I can tell the computer is
still functioning in some way. The screen just won't wake up.

And nothing is logged in the system log around the time of the wake
event?

It's possible one of the peripherals you have connected to the machine
is not handling the wake event properly. You could rule that out by
disconnecting all peripherals before letting the machine sleep. If the
problem happens anyway, you've ruled out peripheral issues.