Stats is preventing sleep on Mojave

[minimum91]

Hi there, thank you for the extremely useful and customisable utility. Is there anything I can do on my end to stop it preventing sleep?

I am running 10.14.6 on 5,1 Mac Pro.

[dehydratedpotato]

@minimum91 I don't know what's going on there but we could try and narrow down what assertions Stats is making to prevent sleep. What is your output from running pmset -g assertions in Terminal?

[minimum91]

@BitesPotatoBacks

Here's the output from the command:

2023-03-09 00:23:34 +0200 Assertion status system-wide: BackgroundTask 1 ApplePushServiceTask 0 UserIsActive 1 PreventUserIdleDisplaySleep 0 PreventSystemSleep 0 ExternalMedia 0 PreventUserIdleSystemSleep 1 NetworkClientActive 0 Listed by owning process: pid 84(apsd): [0x0000419a000ba625] 00:00:17 ApplePushServiceTask named: "com.apple.apsd-waitingformessages-push.apple.com" pid 107(hidd): [0x0000372f00098940] 00:00:00 UserIsActive named: "com.apple.iohideventsystem.queue.tickle.4294976813.3" Timeout will fire in 900 secs Action=TimeoutActionRelease pid 285(UserEventAgent): [0x0000415d000ba54c] 00:01:18 BackgroundTask named: "eu.exelban.Stats.Network.IP" Created for PID: 14032. pid 333(sharingd): [0x000041150001a479] 00:02:30 PreventUserIdleSystemSleep named: "Handoff" Kernel Assertions: 0x10c=USB,BT-HID,MAGICWAKE id=500 level=255 0x4=USB mod=01/01/1970, 03:00 description=com.apple.usb.externaldevice.fd300000 owner=Duet USB id=502 level=255 0x4=USB mod=01/01/1970, 03:00 description=com.apple.usb.externaldevice.1d100000 owner=flirc id=503 level=255 0x4=USB mod=01/01/1970, 03:00 description=com.apple.usb.externaldevice.1a200000 owner=IOUSBHostDevice id=505 level=255 0x4=USB mod=01/01/1970, 03:00 description=com.apple.usb.externaldevice.1a230000 owner=USB Receiver id=506 level=255 0x100=MAGICWAKE mod=08/03/2023, 23:50 description=en0 owner=en0 id=507 level=255 0x100=MAGICWAKE mod=08/03/2023, 23:46 description=en2 owner=en2 id=534 level=255 0x8=BT-HID mod=01/01/1970, 03:00 description=com.apple.driver.IOBluetoothHIDDriver owner=AppleHSBluetoothDevice id=535 level=255 0x8=BT-HID mod=01/01/1970, 03:00 description=com.apple.driver.IOBluetoothHIDDriver owner=AppleHSBluetoothDevice Idle sleep preventers: IODisplayWrangler

[dehydratedpotato]

Well the BackgroundTask assertion is clearly 1 (true), which corresponds with this line:

pid 285(UserEventAgent): [0x0000415d000ba54c] 00:01:18 BackgroundTask named: "eu.exelban.Stats.Network.IP"
Created for PID: 14032.

From what I recall, Stats communicates with exelban's server in order to retrieve user IP, and this could be what's blocking sleep....

@minimum91 do you have the network module enabled in Stats? Perhaps disabling that module could unblock sleep

[exelban]

As I knot network task could not prevent sleep in the macOS.

[exelban]

Yeap. there you have a process that wakes up your mac: pid 333(sharingd): [0x000041150001a479] 00:02:30 PreventUserIdleSystemSleep named: "Handoff"

There is nothing in the logs you shared saying that Stats prevents mac from sleeping.

[minimum91]

Well the BackgroundTask assertion is clearly 1 (true), which corresponds with this line:

pid 285(UserEventAgent): [0x0000415d000ba54c] 00:01:18 BackgroundTask named: "eu.exelban.Stats.Network.IP"
Created for PID: 14032.

From what I recall, Stats communicates with exelban's server in order to retrieve user IP, and this could be what's blocking sleep....

@minimum91 do you have the network module enabled in Stats? Perhaps disabling that module could unblock sleep

Seems to be off, if this is what you are referring to. I just use one RAM block, that's all.

Yeap. there you have a process that wakes up your mac: pid 333(sharingd): [0x000041150001a479] 00:02:30 PreventUserIdleSystemSleep named: "Handoff"

There is nothing in the logs you shared saying that Stats prevents mac from sleeping.

@exelban I know it's a bit counter-intuitive, but Handoff has always been there on my machine, yet it never prevented the Mac from sleeping. As soon as I quit Stats, the machine goes to sleep as normal.

[exelban]

I have running Stats on a few macs and never cause the problem with sleep. Also, I'm pretty sure that the network request could not prevent mac from sleeping.

If you have disabled the network module it means only one: the request was never done. Just the task was registered. So Stats was not even sent a request.

[dehydratedpotato]

@exelban

There is nothing in the logs you shared saying that Stats prevents mac from sleeping.

Actually, there is. See these two lines: BackgroundTask 1 and pid 285(UserEventAgent): [0x0000415d000ba54c] 00:01:18 BackgroundTask named: "eu.exelban.Stats.Network.IP" Created for PID: 14032.

These denote that Stats is creating BackgroundTask named "eu.exelban.Stats.Network.IP" that is preventing sleep (as you can see that BackgroundTask is set to true)

If you have disabled the network module it means only one: the request was never done. Just the task was registered. So Stats was not even sent a request.

I've tried to investigate the source code and I think this isn't occurring from the network access (which like you said wouldn't occur when the module is disabled), but rather the activity of the NSBackgroundActivityScheduler. From my test, it looks as if the schedule closure for ipUpdater runs in the background even with the Network module disabled. So perhaps the very existence of the ipUpdater BackgroundTask could be causing this sleep issue? It's the only connection I can find to "eu.exelban.Stats.Network.IP"...

[exelban]

@BitesPotatoBacks so you want to say that NSBackgroundActivityScheduler prevents sleep?

[dehydratedpotato]

@exelban Yes based on what I could test that seems to be the case... Though, this sleep problem isn't an issue on my Mac Mini (which is running Ventura) so perhaps this sleep prevention is just a problem with the way Mojave might handle background tasks?

[exelban]

It's very strange. If you will debug why it's happening and how it could be fixed ping me. I cannot reproduce that bug on a macbook or mac studio. Also before have a hackintosh, and even there I have no problems with sleep.

[dehydratedpotato]

@exelban perhaps it could be fixed by invalidating the ipUpdater when the Network module is disabled, and then rescheduling with setIPUpdater() on re-enable? You could make the enable() and disable() Module methods open, that way you could override them in the Network subclass and do the invalidation/rescheduling from within those overrides.

I believe that could solve this problem, but like I said, I cannot recreate the sleep prevention, I only observed the persistence of the background activity. So If those changes were made, @minimum91 would have to test the new build in order to validate my guessing. 🤔

[exelban]

Stats already has it. Will be available in the next release