So far so good - fingers crossed! Thanks for making the suggestions.
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.
Battery drain issue while MacBook Pro was sleeping
- Thread starter chris4565
- Start date
- Sort by reaction score
Can you provide me with the terminal instructions for doing this changes? I'm also confronting these issues on my early 2015 MacBook Air since updating to macOS 10.15.6. The weird thing is that I have had this notebook for four years and all the way up to 10.15.5 sleep mode with lid closed worked fine.
First, in Terminal, pmset -g to display and take note of your current settings.
Then,
sudo pmset standbydelayhigh 3600return, enter User pw, and return. Then
sudo pmset standbydelaylow 3600and return.
Use whatever value you want. 3600 is one hour. 10800 is the formerly standard (for MBAs and retina MBPs) 3 hours.
It's weird. I have a 2017 MBP TouchBar 16/512 that worked great until a few weeks ago. And then the battery started draining to 0 while sleeping in just 6-7 hours. I keep the system and apps up to date.
I did a thousand things, from disconnecting from iCloud to uninstalling all apps, to shutting down WIF/BT/notifications etc. I finally created a new volume and installed Big Sur Release Candidate this week, completely new setup. Alas, same issue. For now, I didn't want to change the delays and hibernation mode as I'm trying to get the MacBook back to the exact same situation as before I had this issue.
When looking at pmset log, the same scheme reproduces every 12-15 sec (see quote below). It looks as though 5 seconds into going to sleep there is a InternalPreventSleep event created with "wait for device enumeration" as the culprit.
Would anyone have encountered this specific problem?
As an addition: I have had an issue with one of the 4 TB ports for months, and I start wondering if my problem is hardware (like motherboard getting worst and worst).
Thanks a lot to all of you spending so much time helping!
I did a thousand things, from disconnecting from iCloud to uninstalling all apps, to shutting down WIF/BT/notifications etc. I finally created a new volume and installed Big Sur Release Candidate this week, completely new setup. Alas, same issue. For now, I didn't want to change the delays and hibernation mode as I'm trying to get the MacBook back to the exact same situation as before I had this issue.
When looking at pmset log, the same scheme reproduces every 12-15 sec (see quote below). It looks as though 5 seconds into going to sleep there is a InternalPreventSleep event created with "wait for device enumeration" as the culprit.
Would anyone have encountered this specific problem?
As an addition: I have had an issue with one of the 4 TB ports for months, and I start wondering if my problem is hardware (like motherboard getting worst and worst).
Thanks a lot to all of you spending so much time helping!
It's a puzzle, and not the fun kind. I just ordered a new battery for my 6s because after reinstall seemed to not fix a sudden total battery failure last night, today the battery works fine. :/
I now am pretty convinced you should be on "standby" RAM-poweroff regimen even keeping the 3 setting, that OS X just treats hibernatemode 3 as "standby" on retina MBPs. In which case the drain is still high. But should be doesn't mean is. If RAM is being powered, that's where your battery is going.
I wish I could see the delay defaults for retina MBPs in High Sierra.
Well, looking at my logs it appears that my Mac wakes every 3hrs on the dot and repeats the logs below about bluetooth. I suppose the next investigation is why. I do have a Apple keyboard and a Logitech mouse that are bluetooth but neither of these is in use at 1am in the morning. By the looks the logs each 3 hrs are the same. Time for google.
Code:
2020-11-12 01:14:12 +1000 Kernel Client Acks Delays to Sleep notifications: [RP12 driver is slow(msg: SetState to 2)(357 ms)] [AirPort_BrcmNIC driver is slow(msg: SetState to 2)(466 ms)] [RP12 driver is slow(msg: SetState to 0)(1208 ms)]
2020-11-12 01:14:12 +1000 AppWakeReason AppWoke:com.apple.bluetoothd-blueavengers Reason:BlueAvengers: Toggle beacon state
2020-11-12 01:14:12 +1000 Assertions PID 256(mDNSResponder) Created MaintenanceWake "mDNSResponder:maintenance" 00:00:00 id:0x0xd0000a5ea [System: No Assertions]
2020-11-12 01:14:12 +1000 Assertions PID 256(mDNSResponder) Released MaintenanceWake "mDNSResponder:maintenance" 00:00:00 id:0x0xd0000a5ea [System: No Assertions]
2020-11-12 01:14:13 +1000 DarkWake DarkWake from Standby [CDN] : due to RTC/Maintenance Using BATT (Charge:67%) 0 secs
2020-11-12 01:14:13 +1000 HibernateStats hibmode=3 standbydelaylow=7200 standbydelayhigh=7200 rd=598 ms
2020-11-12 01:14:13 +1000 WakeTime WakeTime: 4.790 sec
2020-11-12 01:14:13 +1000 Kernel Client Acks Delays to Wake notifications: [AirPort_BrcmNIC driver is slow(msg: SetState to 2)(463 ms)] [UPSB driver is slow(msg: SetState to 2)(540 ms)] [UPSB driver is slow(msg: SetState to 2)(540 ms)] [IOBluetoothHostControllerUARTTransport driver is slow(msg: SetState to 1)(636 ms)]
2020-11-12 01:14:13 +1000 Sleep Entering Sleep state due to 'Maintenance Sleep':TCPKeepAlive=disabled Using Batt (Charge:67%) 10803 secs
2020-11-12 01:14:14 +1000 Wake Requests [*process=bluetoothd request=Maintenance deltaSecs=10798 wakeAt=2020-11-12 04:14:12 info="com.apple.bluetoothd-blueavengers"]
2020-11-12 01:14:14 +1000 PM Client Acks Delays to Sleep notifications: [com.apple.apfsd is slow(267 ms)] [PLSleepWakeAgent is slow(427 ms)] [systemstats is slow(876 ms)] [com.apple.duetactivityscheduler is slow(1295 ms)]Well googling seems to imply that the iBeacon and the avenger item is related to find my Mac? I've never done anything with Find My Mac and seems this is now just "Find My" instead. I've tried setting the autopoweroffdelay to a low value to see if that makes a difference. Open to any ideas that others have tried that may have seen this previously.
ok I had to read pmset man again and correct my first reply.
I would try hibernatemode 25 or just shut down the machine every night before resorting to autopoweroff, since you wouldn't really be "fixing" sleep at that point
I would try hibernatemode 25 or just shut down the machine every night before resorting to autopoweroff, since you wouldn't really be "fixing" sleep at that point
Well even with tcpkeepalive set to 0, and hibernate mode set to 25, my MBP continued to used 6% over night in 7 hrs. Its still waking up due to the bluetooth by the looks and whatever is blue avenger. I spent an hour googling yesterday for that and can not find a reason. Tonight I'll try disabling bluetooth on the MBP before it goes to sleep. I'm not sure what its doing or why but the major difference to last night is that the bluetooth regular woke it every 3 hrs and now it seems every hour.
It definitely seems to be that it is meant to be waking by the line.
Wake Requests [*process=bluetoothd request=Maintenance deltaSecs=3597 wakeAt=2020-11-13 03:33:08 info="com.apple.bluetoothd-blueavengers"] [process=powerd request=AdaptiveWake deltaSecs=9703 wakeAt=2020-11-13 05:14:54]
However a pmset -g sched states there is nothing scheduled. More googling I suppose.
It definitely seems to be that it is meant to be waking by the line.
Wake Requests [*process=bluetoothd request=Maintenance deltaSecs=3597 wakeAt=2020-11-13 03:33:08 info="com.apple.bluetoothd-blueavengers"] [process=powerd request=AdaptiveWake deltaSecs=9703 wakeAt=2020-11-13 05:14:54]
However a pmset -g sched states there is nothing scheduled. More googling I suppose.
Code:
2020-11-13 02:33:08 +1000 Kernel Client Acks Delays to Sleep notifications: [IOBluetoothHostControllerUARTTransport driver is slow(msg: SetState to 0)(508 ms)]
2020-11-13 02:33:08 +1000 AppWakeReason AppWoke:com.apple.bluetoothd-blueavengers Reason:BlueAvengers: Toggle beacon state
2020-11-13 02:33:08 +1000 Assertions PID 243(mDNSResponder) Created Maint
enanceWake "mDNSResponder:maintenance" 00:00:00 id:0x0xd0000866c [System: DeclUser
kDisp]
2020-11-13 02:33:08 +1000 Assertions PID 243(mDNSResponder) Released Main
tenanceWake "mDNSResponder:maintenance" 00:00:00 id:0x0xd0000866c [System: DeclUser kDisp]
2020-11-13 02:33:09 +1000 DarkWake DarkWake from Standby [CDN] : due to RTC/Maintenance Using BATT (Charge:68%) 0 secs
2020-11-13 02:33:09 +1000 HibernateStats hibmode=25 standbydelaylow=7200 standbydelayhigh=7200 rd=232 ms
2020-11-13 02:33:09 +1000 WakeTime WakeTime: 3.998 sec
2020-11-13 02:33:09 +1000 Kernel Client Acks Delays to Wake notifications: [UPSB driver is slow(msg: SetState to 2)(539 ms)] [UPSB driver is slow(msg: SetState to 2)(538 ms)] [RP12 driver is slow(msg: SetState to 1)(1200 ms)]
2020-11-13 02:33:09 +1000 Sleep Entering Sleep state due to 'Maintenance Sleep':TCPKeepAlive=disabled Using Batt (Charge:68%) 3602 secs
2020-11-13 02:33:09 +1000 AppWakeReason AppWoke:com.apple.bluetoothd-blueavengers Reason:BlueAvengers: Roll beacon advertisement
2020-11-13 02:33:11 +1000 Wake Requests [*process=bluetoothd request=Maintenance deltaSecs=3597 wakeAt=2020-11-13 03:33:08 info="com.apple.bluetoothd-blueavengers"] [process=powerd request=AdaptiveWake deltaSecs=9703 wakeAt=2020-11-13 05:14:54]
2020-11-13 02:33:11 +1000 PM Client Acks Delays to Sleep notifications: [com.apple.apsd is slow(1938 ms)]Well I'm thinking this has something to do with "Find My". According to here these beacons are provided by having multiple devices on your apple account (I do) and that "Find My" is enabled on one. They swap encryption keys and then regularly send beacons to locate the other device. Receiving Apple devices take the public key and send back to Apple. Decryption is only possibly if you have a second device that had the original private key from initial connection on your devices. So while my current MBP has "Find My" disabled my other Mac Air had it enabled. I'm guessing that the beacons from my laptop are sent as a result because both devices are in the same account. Tonight will tell I guess.
Wired article on how "Find My" works was good.
Wired article on how "Find My" works was good.
I'm hopeful i have solve the avenger bluetooth wake up issue. Only time will tell. "Find My" was turned off on my Mac. Or at least the icloud tick was not on the capability. However, if i selected my MBP on the left hand side within the accounts settings it stated that it was on and gave it a green led. I could enable and disable in icloud and that green led would not change. Now, a while ago, because of Catalina, Apple changed my logic board, and my old MBP was still in the account. ie. I had 2 MBP in there as devices. Both of these had the same serial number. So out of frustration i deleted the device that i thought was no longer my current device (obviously no visible difference between the two) and as soon as i did that the green led against "Find My" disappeared!!! So I am assuming that "Find My" was enabled on my old logic board before Apple replaced it and some how that setting seemingly makes its way to my new logic board but for some reason never synchronises with the software settings. I suppose I will find out in a few hours but I think it looks promising.
So yes...my issues where "Find My" and some thing to do with Apple rather than my local machine. Delete all machines in my account, even those that do not exist anymore and then everything sleeps as expect. Its still using 5% of battery in 7hrs of sleep but at least now I know its not waking up every few hours. Thanks for the inputs much appreciated.
For those who haven't thought about it, I was having this same battery problem while under the last two point updates to Catalina and once I upgraded to Big Sur the problem has been resolved. Not ideal if you depend on apps that haven't been updated to be compatible with Big Sur yet but it's worth a try to see if it fixes the battery drainage while in sleep mode. Hope it helps.
Hello, I can't understand how hibernation works. BTW I have 13 MBP 2020 and latest Catalina.
First of all, a little story. I had my battery fully drained while my MBP was in bag during the last weekend (just 2 days in closed lid mode). So.. I started the investigation
Here what I've found in pmset log
The last line is the time when I've opened the lid. The battery was empty.
I've tried to analyze istats charts and syslog entries to find what was going on, but it seems that nothing was happening. The last istat chart point (cpu usage, memory, battery percentage, etc) was on 28.11.20, before the weekend. Also, syslog was empty.
But the battery drained somehow during sleep...
Yesterday I've set some pmset settings:
and my battery pmset -g looked like that:
System-wide power settings:
VACTDisabled 0
Currently in use:
standby 1
standbydelaylow 3600
halfdim 1
hibernatefile /var/vm/sleepimage
proximitywake 0
powernap 0
gpuswitch 2
disksleep 15
standbydelayhigh 10800
sleep 30 (sleep prevented by sharingd)
hibernatemode 3
ttyskeepawake 1
displaysleep 10
tcpkeepalive 0
highstandbythreshold 50
acwake 0
lidwake 1
So, I was expecting that my MBP will hibernate in an hour after I close the lid. But when I've opened the lid it started immediately (just like RAM was not powered off).
BTW, batter drain during 9 sleep hours was 2%.
First of all, a little story. I had my battery fully drained while my MBP was in bag during the last weekend (just 2 days in closed lid mode). So.. I started the investigation
Here what I've found in pmset log
2020-11-28 05:44:41 +0300 Sleep Entering Sleep state due to 'Clamshell Sleep':TCPKeepAlive=active Using Batt (Charge:77%) 7 secs
2020-11-28 05:44:48 +0300 DarkWake DarkWake from Deep Idle [CDN] : due to EC.ARPT/Maintenance Using BATT (Charge:77%) 46 secs
2020-11-28 05:45:34 +0300 Sleep Entering Sleep state due to 'Maintenance Sleep':TCPKeepAlive=inactive Using Batt (Charge:77%) 7155 secs
2020-11-28 07:44:49 +0300 DarkWake DarkWake from Deep Idle [CDN] : due to EC.SleepTimer/SleepTimer Using BATT (Charge:74%) 0 secs
2020-11-28 07:44:49 +0300 Sleep Entering Sleep state due to 'Maintenance Sleep':TCPKeepAlive=inactive Using Batt (Charge:74%) 76 secs
2020-11-28 07:46:05 +0300 DarkWake DarkWake from Standby [CDN] : due to EC.SleepTimer/HibernateError Using BATT (Charge:74%) 46 secs
2020-11-28 07:46:51 +0300 Sleep Entering Sleep state due to 'Maintenance Sleep':TCPKeepAlive=inactive Using Batt (Charge:74%) 14290 secs
2020-11-28 11:45:01 +0300 DarkWake DarkWake from Deep Idle [CDN] : due to EC.RTC/SleepService Using BATT (Charge:69%) 0 secs
2020-11-28 11:45:01 +0300 Sleep Entering Sleep state due to 'Sleep Service Back to Sleep':TCPKeepAlive=inactive Using Batt (Charge:69%) 21641 secs
2020-11-28 17:45:42 +0300 DarkWake DarkWake from Deep Idle [CDN] : due to EC.RTC/Maintenance Using BATT (Charge:63%) 45 secs
2020-11-28 17:46:27 +0300 Sleep Entering Sleep state due to 'Maintenance Sleep':TCPKeepAlive=inactive Using Batt (Charge:63%) 34113 secs
2020-11-28 17:46:30 +0300 Assertions Summary- [System: DeclUser kDisp] Using Batt(Charge: 63)
2020-11-29 03:15:00 +0300 DarkWake DarkWake from Deep Idle [CDN] : due to EC.RTC/SleepService Using BATT (Charge:47%) 0 secs
2020-11-29 03:15:00 +0300 Sleep Entering Sleep state due to 'Sleep Service Back to Sleep':TCPKeepAlive=inactive Using Batt (Charge:47%) 63906 secs
2020-11-29 21:00:06 +0300 DarkWake DarkWake from Standby [CDN] : due to EC.RTC/SleepService Using BATT (Charge:47%) 2 secs
2020-11-29 21:00:08 +0300 Sleep Entering Sleep state due to 'Sleep Service Back to Sleep':TCPKeepAlive=inactive Using Batt (Charge:18%) 55153 secs
2020-11-30 12:19:18 +0300 Assertions Summary- [System: DeclUser kDisp] Using AC(Charge: 0)
The last line is the time when I've opened the lid. The battery was empty.
I've tried to analyze istats charts and syslog entries to find what was going on, but it seems that nothing was happening. The last istat chart point (cpu usage, memory, battery percentage, etc) was on 28.11.20, before the weekend. Also, syslog was empty.
But the battery drained somehow during sleep...
Yesterday I've set some pmset settings:
global:
sudo pmset -a disksleep 15
charger:
sudo pmset -c sleep 120
sudo pmset -c powernap 1
sudo pmset -c displaysleep 20
battery:
sudo pmset -b tcpkeepalive 0
sudo pmset -b sleep 30
sudo pmset -b displaysleep 10
sudo pmset -b standbydelayhigh 10800
sudo pmset -b standbydelaylow 3600
sudo pmset -b proximitywake 0
sudo pmset -b powernap 0
and my battery pmset -g looked like that:
System-wide power settings:
VACTDisabled 0
Currently in use:
standby 1
standbydelaylow 3600
halfdim 1
hibernatefile /var/vm/sleepimage
proximitywake 0
powernap 0
gpuswitch 2
disksleep 15
standbydelayhigh 10800
sleep 30 (sleep prevented by sharingd)
hibernatemode 3
ttyskeepawake 1
displaysleep 10
tcpkeepalive 0
highstandbythreshold 50
acwake 0
lidwake 1
So, I was expecting that my MBP will hibernate in an hour after I close the lid. But when I've opened the lid it started immediately (just like RAM was not powered off).
BTW, batter drain during 9 sleep hours was 2%.
Unclear on how long until you opened the lid? When over 50% battery, pmset will use your 3 hour standbydelayhigh setting.
If you're getting instant wakes after 9 hrs, you can try forcing hibernate by setting hibernatemode to 25, but 2% per 9 hours isn't what killed your battery anyway.
It was that bizarre holding at 47% and then giant drop out of nowhere.
I would keep an eye on what your battery does next time it gets below 50, whether during sleep or not
Thanks for the reply.
Well, I don't think it's battery issue, because:
1. I can easily get 6-7 hours of display on time while on battery and there is no sudden percentage drops.
2. The battery was manufactured 6 months ago
I'll try to explain what is concerning me. Let's talk about sleep process.
But first, let's assume there is no software running or hardware connected that is preventing sleep). I mean we have ideal circumstances.
Here is my pmset -g (while running on battery):
Battery Power:
lidwake 1
standbydelayhigh 10800
proximitywake 0
standby 1
standbydelaylow 10800
ttyskeepawake 1
highstandbythreshold 50
powernap 0
gpuswitch 2
hibernatefile /var/vm/sleepimage
hibernatemode 3
displaysleep 10
sleep 30
tcpkeepalive 0
halfdim 1
acwake 0
lessbright 1
disksleep 15
Let's pretend I use my machine with no charger plugged.
At 7:00pm I've done my work and went somewhere without closing the lid (macbook left on the battery).
At 7:10pm my display should turn off (according to displaysleep 10 setting)
At 7:30pm MBP should go to sleep mode with RAM still powered (according to sleep 30 setting)
At 10:30pm MBP should go into standby mode (hibernate) and power off ram (according to standbydelayhigh/standbydelaylow 10800 setting - they are identical and standby 1 setting)
...
On the next day at 11:00am (more than 12 hours have passed since I touched my mac last time) I press any button and I expect to see my mac waking a bit longer, because it has to read SSD to get RAM snapshot (maybe progress bar should appear).
Is this how it is supposed to be?
...
But real life shows that mac doesn't go to standby (hibernate). It wakes immediately the next morning.
Maybe I miss something?
Maybe the case is hibernatemode 3 setting? But as far as I understand it should not prevent standby (hibernate) - it's default value is 3 on new machines.
Maybe MBP goes into standby and I dont notice it? Yes, SSDs are fast nowadays, but not that much... I have more than 20GB RAM used (on 32GB machine) and such a huge snapshot cannot be read from SSD in a moment.
Is there is any log entry showing that MBP went into standby (hibernate, with RAM powered off)?
Right, I think your RAM is staying powered on because of a change to SMC behavior pushed in Mojave and Catalina this year, but since pmset isn't actually reflective of what SMC is going to do with RAM in retina MBPs, who can say. I haven't seen a report of a retina MBP user with Mojave or Catalina still seeing the slow wake that reflects RAM being powered off, but maybe some are... Either way your RAM is clearly staying powered on, which will use battery.
Right, I wouldn't jump to conclude anything is wrong with the battery because of one weird sudden drop, but I would keep an eye out.
Hi, sorry to bring up the old thread. Is there any workable solution for this? I did turn off Find My, tcpkeepalive 0, and set standbydelaylow & high to 3600, but there are still wake and sleep below every 2 minutes:
Wake from Standby [CDNVA] : due to EC.RTC/Alarm Using BATT
Entering Sleep state due to 'Clamshell Sleep':TCPKeepAlive=disabled Using Batt
In the end, the battery eventually went to 0%. Mine is MBA 2020 / Catalina.
Thanks!
Wake from Standby [CDNVA] : due to EC.RTC/Alarm Using BATT
Entering Sleep state due to 'Clamshell Sleep':TCPKeepAlive=disabled Using Batt
In the end, the battery eventually went to 0%. Mine is MBA 2020 / Catalina.
Thanks!