Q: rMBP 2015 Wake Reason: ARPT (Network)

The reason why I infer it could be related to the router is that I've put my mac to sleep in several different routers. And for one router it would wake every ten minutes or so. But for another one, it waked up randomly without obvious interval. After all, Apple should be responsible for this issue somehow.

I have found out the same behaviour. Currently, I am visiting my parents, and my rMBP has no "sleep problems" while connected to the WiFi router here. As soon as it is connected to my router at home, it wakes up every 10-20 minutes. The only thing that has changed is the router.

I'm still wondering why this isn't happening on the older macs than 2015? I bet it has something to do with the HW on newest models or HW <> SW bug in OS X. IS there any chance that continuity would cause this? Have anyone disabled continuity permanently and ARPT wake ups still happen?

And best question.. is this happen on ALL 2015 macs? I bet it's not.

P.S. It was now 4th time when my rMBP 15 was crashed during night from sleep whole night ARPT wake ups till 05AM and then bang it was crashed, fans on max and superior hot. Lets see if my replacement device (what will be delivered next week) will have same issue, if not then it must have something to do with HW.

Received the replacement unit today and configured it on the office. Right away when I arrive to home ARPT wake ups started. Timestamp matches.

What I wonder does this "issue" have something to do with AirPort Extreme (0x14E4, 0x133) and Broadcom BCM43xx (7.15.166.24.3). My old unit had same Broadcom AirPort card just like this replacement device. I sent email to the Senior that they should be in touch with their engineers ASAP for solving this problem.

Mine is Broadcom BCM43xx 1.0 (7.21.94.23.1a20)

In conclusion, the new unit you got doesn't solve the problem, right? I strongly doubt it's a software issue.

Exactly, this is what I've been observing!

The first night after installation of El capitan public beta, my rMBP drained up to 70%+ of battery, Lucky I opened it before the battery drained completely (at 4%), which would harm the battery I think. Even though the ARPT wake problem existed before but it was very mild, never more than 15% per night, 70%+ is completely unacceptable!

And I find that time machine, after confirming on a dialogue, has deleted all my old backups and built a brand new one! Looks there's no way back to Yosemite.

I have one theory in mind though I didn't know whether it was appropriate to post it here in Apple forum, but obviously I have to say it now and see if anyone breaks it: My Time Machine, which I think is what the "ARPT" refers to, is in fact an afp volume on a home based Ubuntu Server, and I have to type a command line in order to make it appear in the time machine volume enumeration list. On the server there's one other afp volume which I used to store documents and movies.

For anyone in this thread that is in similar situation, my server OS is ubuntu server 14.04, the package to convey the afp service is netatalk. Bonjour service is also configured using avahi (but I'm not sure if it is actually working).

There's an irresistible reason for my theory : the Macbook behaves absolutely normally around other wireless environments, the problem appears whenever I get home and the macbook gets access to the file server.

If my theory holds then this ARPT wake problem has involved third party hardware and deep command-line-sudo-level customisation by the user. Then this is a good reason why Apple is not answering and not investigating into this problem : it's been around for more than 3 months and it has gone through into El Capitan.

If my theory holds then sadly we're on our own way of solving the problem, abandon the custom-made capsule or debug the problem oneself!

If there's anyone who's using solely apple manufactured equipments and hasn't done anything in command line to modify the OS behaviour but also suffers this problem, then we can hope to see it be resolved one day.

I'm sorry but this have nothing to do with command-line customisation or DIY time machine backup things. With factory image and zero backups done via time machine ARPT wake ups happen. I'll push my contacts (Senior Advisors) in Apple for solving this thing up with their engineers. I already sent email for these 2 guys on friday so I'm expecting answer on tomorrow.

Me too, absolutely stock out of the box rMBP with Yosemite does this.  It has nothing to do with any customization.  The most frustrating thing is that we've clearly identified the problem but can't get any response from Apple.  Apple needs to have a process for elevating problems like these to engineers straight from this forum and then needs to keep us informed on progress.  The complete silence is extremely annoying. 

That's a big relief for me actually! Because I rely heavily on my ubuntu file server, good to know it's not the villain.

 
The problem is mystifying again though, for one thing I have two routers of completely different brands and different grades both causing this problem which is over coincidence for me; for another apple must have been very hard working hiding themselves from this thread in order to ignore such a big and severe problem.

 
If this kind of problem persists for long enough it could be a reason people abandon apple computers, after all a big reason I chose apple over microsoft and a biggest difference of apple over microsoft is that for apple I never have to worry about what or how much I've been doing or working on and I could close the lid immediately whenever I want and go on a vacation for several days then go back exactly where I left simply by reopening the lid, in the case of windows before I leave work I have to reflect about what processes have been running in the background, whether it's keeping the computer awake, close some big memory killer programs, and think about the safety of putting the computer to sleep and then at last give the computer time for gracefully entering the sleep, moreover I have to worry about if the computer will die in its sleep and I'd probably decide finally to shut it down before I actually go on vacation. Looks now I have to do the same with macs.

 
Sorry I have been over-wording, in short I think reliability is what defines apple apart from microsoft. My iPad 3 has been up and running for more than 3 years and never died for once, for example. New apple products are seeing more and more experience-demolishing bugs, and I hope I'm not over-minded and this doesn't indicate a landslide of apple.

 
I hope apple could see this post and work harder on a fix.....

I am also experiencing the same behavior on a 15-inch rMBP from mid-2015.   First wake reason is "Wake on TCP Timeout" and then it repeats with "Wake on excessive wake" until the battery dies.  It has been doing this since I purchased the laptop and is still doing it after upgrading to the El Capitan RC--no resolution in sight, but I have contacted Apple support and have a follow-up call scheduled for tomorrow morning.  I'll be sure to refer to this thread and mention a number of new out of the box laptops are experiencing the same battery drain when connected to certain WIFI routers (mine appears to only happen while at home on my newer Arris cable modem/router from TimeWarner Cable).  I'll post back if I find anything helpful...until then I'll try to just turn off my wifi.

Sep 10 20:09:50 Roberts-MacBook-Pro kernel[0] <Notice>: ARPT: 714.313518: ARPT: Wake Reason: Wake on TCP Timeout

Sep 10 20:09:52 Roberts-MacBook-Pro kernel[0] <Notice>: ARPT: 716.293141: ARPT: Wake Reason: Wake on TCP Timeout

Sep 10 20:09:52 Roberts-MacBook-Pro kernel[0] <Notice>: [HID] [ATC] AppleDeviceManagementHIDEventService::processWakeReason Wake reason: Host (0x01)

Sep 10 20:15:37 Roberts-MacBook-Pro kernel[0] <Notice>: Wake reason: ?

Sep 10 20:15:37 Roberts-MacBook-Pro kernel[0] <Notice>: ARPT: 750.573748: ARPT: Wake Reason: Wake on excessive wake

Sep 10 20:15:40 Roberts-MacBook-Pro kernel[0] <Notice>: ARPT: 753.952697: ARPT: Wake Reason: Wake on excessive wake

Sep 10 20:15:40 Roberts-MacBook-Pro kernel[0] <Notice>: [HID] [ATC] AppleDeviceManagementHIDEventService::processWakeReason Wake reason: Host (0x01)

Sep 10 20:16:55 Roberts-MacBook-Pro kernel[0] <Notice>: Wake reason: ?

-R

Sure enough, disabling wifi on my laptop before going to bed resulted in a happy battery this morning.  The peculiar behavior of certain wifi networks causing this issue is also present as I only have the issue when connected at home, but at the office the wifi networks do not cause the sleep issue.

What is the best way to report this software bug to Apple?  Should I follow through with their phone support or do I just wait until they finally figure it out like I did with my 2012 rMBP video driver issue?

-R

I spoke with Apple phone support and escalated this issue--they linked to this thread and sent it up the chain to their network engineering group.

Good to know. Hope they really pay due attention to this problem and solve it.

Today I put my MacBook 12" to sleep (closed the lid) and captured the Wifi communication between it and my router. Looking at the communication with Wireshark I noticed that the first thing the MacBook does after the unwanted wake up is sending TCP packets to IP 17.172.239.3 on port 5223.

After some searching with Google it seems this has something to do with the imagent App which is in some way responsible for FaceTime.

Can this be correlated with the wake up and perhaps even the "ARPT: Wake Reason: Wake on TCP Timeout" message in syslog?

Does anybody here have FaceTime disabled and is still experiencing the unwanted wake ups?

(If an Apple technician wants to have a look at it, I can gladly provide the pcap file with the logged Wifi communication. Just contact me in case!)