Apple TV 4K - Ethernet constantly connects/disconnects while in sleep mode

Question

Level 1

30 points

Apple TV 4K - Ethernet constantly connects/disconnects while in sleep mode

I have been running down an issue where my Apple TV 4K, when connected via ethernet, does NOT maintain a connection link UP status while in sleep mode; it cycles up and down several times per minute. If the ATV4K is awake and playing content, this behavior does not happen and all is well. Most people probably wouldn't know or care about this, but I do since it causes unnecessary network chatter and is generally disruptive network behavior. I also have a 4th Gen Apple TV that is on the same ethernet switch and does NOT exhibit this behavior while sleeping, so I'm quite confident that this is isolated to the ATV4K only. I called Apple Support to report the issue, but they had no knowledge of the behavior.

Can anyone else confirm this behavior with their Apple TV 4K? I'm wondering if I have a defective unit and if I need to request an RMA, or if this is a problem with the ethernet controller or driver. That's what I'm hoping someone else in the Apple Community can confirm/deny.

Data points from troubleshooting

- Ethernet connection constantly disconnects / reconnects in sleep mode ONLY

- When the link status cycles, the ethernet connection negotiates alternately from 100Mbps to 1Gbps (could be part of the problem and suggests software/driver issue to me)

- WiFi connection does NOT exhibit the same disconnect/reconnect behavior in sleep mode

- Apple TV 4th Gen does NOT exhibit this behavior on ethernet (connected to same switch/router)

- Problem persists after bypassing the installed ethernet cabling and switch and connect directly to the router (several different ports)

- iOS 11.1 update had no effect

- DHCP lease renewals are not the root cause, but a symptom of the link status going down/up

- Hard-coding IP address only eliminates DHCP leasing, not the problem itself

And before anyone suggests it, I am NOT willing to leave the ATV4K on all the time to prevent this, and I am not switching to WiFi when I have a perfectly good ethernet connection. I am interested in prompting Apple into either replacing a defective unit or resolving the underlying issue because this is undesirable behavior from a network engineering perspective.

Thanks,

Andy

Apple TV 4K, iOS 11.1

Posted on Nov 7, 2017 9:20 AM

Reply

Answer 1

Mar 22, 2018 1:52 PM in response to nickguy

I can confirm this is still a problem. I had the problem with the 4th gen, and when I upgraded to 4K model. I switched to a new direct cable and moved ports, problem persists. I have a Cisco SG500X, latest firmware, and it is the only device that floods the log with link up/link down as often as every 30 secs to 3-5 mins. I opened a ticket, uploaded the requested logs, they came back with basically "we see nothing wrong, live with it". Which to be fair, it is just a big annoyance, not a show stopper. For now, I have sleep set to "never" which fixes the issue.

I moved the 4th gen Apple TV to a different room that is connected to a Cisco Catalyst 2960G and a different reciever/TV. The 4th gen unit no longer goes up and down constantly when sleeping. I'm thinking it may have something to do with the HDMI triggering something or some HDMI chatter on the other TV. I guess the next test would be to turn sleep back on and unplug the HDMI then watch the switch log.

I wish they would just fix the bug.

I was told that the best thing to do is to go here, https://www.apple.com/feedback/ , and put in a feature request. Support suggested that if enough people did this then it might get fixed. lol.

Jason

Reply

Answer 2

Andypeppa Author

Level 1

30 points

Nov 7, 2017 9:54 AM in response to Diana.McCall

Hi Diana,

I have already opened a case with Apple on this issue. This post is to ask the community at large to help validate this behavior and report if there are more instances of this problem out in the world. Apple could then add that information as a reproducible data point to investigate if this is a more pervasive issue with the ethernet controller / driver implemented in the Apple TV 4K.

Reply

Answer 3

Nov 7, 2017 10:17 AM in response to Andypeppa

Good. I hope you can get some resolution. You're certainly not the only one to experience problems with ethernet. Just search this forum (Using Apple TV) for 4K and ethernet. I personally experienced the repeated random DHCP requests problem with my Apple TV 4 on ethernet. My personal theory is that tvOS is based on iOS, which does not include ethernet support. So the ethernet driver in tvOS is a bit of an orphan.

Reply

Answer 4

Andypeppa Author

Level 1

30 points

Nov 7, 2017 10:27 AM in response to Diana.McCall

The excessive DHCP requests is how I first ID'ed the issue in my environment, but as I dug further, I found that they were merely a consequence of the ethernet link status bouncing up & down. The fact that you saw the same excessive DHCP requests on your ATV4 is an interesting point since I also have one of those but have not seen the same behavior with it. That would indicate to me that the issue is something more transient than a poorly-coded driver, which I would expect to be more or less consistent (assuming that the ethernet controller is same across the ATV4, which I highly doubt b/c Apple is known to have different hardware sets for the same product). I definitely would expect the ATV4K to have a different ethernet controller since it is a 1Gbps controller whereas the ATV4 was only a 100Mbps controller.

I am going to let this thread soak for a few more days to allow others to test/confirm/deny, then I will re-open my case and ask about the possibility of a replacement to eliminate a singular hardware fault.

Thanks for adding to my post!

Reply

Answer 5

Nov 10, 2017 12:51 PM in response to Andypeppa

Hello! I have my Apple TV 4K plugged into an HPE 1920-24 gigabit switch. When the Apple TV is asleep, it fills the switch log with port up/down messages. I can also confirm that for me at least, I did not see this behavior with the Apple TV 4th Gen. I have been trying different port settings on the switch to see if any made a difference but have not found a solution to this point.

Reply

Answer 6

Andypeppa Author

Level 1

30 points

Nov 10, 2017 12:56 PM in response to AspenWiedii

Hi @AspenWiedii,

Thanks for chiming in here. I'm not happy with the issue itself, but it's nice to know that someone else is seeing the same behavior. I will re-open my case with this information and see if we can get some visibility into this by Apple's engineering team. It's also worth noting that I am using a Cisco SG200-26 switch from the small business line.

I'll also mention to you that if you are bent on eliminating the behavior, the ATV4K WiFi connection does not suffer from the same disconnect/reconnect behavior. This, of course, is a workaround and not a resolution...hopefully that will follow soon if we can get enough visibility into this issue.

Reply

Answer 7

opragel

Level 1

4 points

Jan 1, 2018 9:29 AM in response to Andypeppa

I'm also seeing the exact behavior you described. tvOS 11.1.

I'm okay with having it never sleep to mitigate, but definitely seems like a bug and just annoying if you're one of the few folks watching the log.

Reply

Answer 8

Jan 3, 2018 2:35 PM in response to Andypeppa

Andy,

I have the same problems as you described. I did everything you did, Andy, plus I configured my cisco switch interface to be either auto negotiate the speed/duplex and hard coded them to 1Gig/Duplex and still have the problem. I even set the apple TV to sleep never, but it still goes to sleep and floods my syslog server with all these flapping interface messages. I bought this 4K TV to replace the previous apple tv. I just swapped them, so the new one was using the same cable and switch port as the old one. The old one didn’t have this issue, so my cable and switch port is fine.

I already took one back to the Apple store and they gave me a new one. I thought it was just that one device was bad, but the new one they swapped me was doing the same thing. That’s when I did an internet search and found this thread you posted.

I just got off the phone with Apple support after being escalated to Tier 2 only to be told that the engineers don’t tell Tier 2 what they’re working on; Ivan said he’ll escalate it to the engineers so that more people complain about this and it gets attention. Ivan told me to keep checking the Apple Web site for IOS upgrade release notes for anything that fixes this issue. I think this is going to take months to resolve.

That said, I’m still in my 30 day return policy and I’m just going to take this back and get my money and probably connect up my old Apple TV even though its incompatible with 4K. And I just bought a new 4K TV to stream 4K movies. Or, I’ll check out RoKu or Amazon stick or something.

I get the impression ALL of the Apple 4K TV devices are faulty. So take yours back or don’t buy it until they fix it.

Reply

Answer 9

Jan 15, 2018 9:09 PM in response to Andypeppa

Same here, got a new Apple TV 4K yesterday and woke up to 200+ e-mails from my monitoring system due to port flapping. Seems the port is going up&down every few seconds.

Doesn't this also affect AirPlay availability?

Very short excerpt from my switch log:

Jan 16 05:58:57 192.168.***.*** ("US8,f09fc2c6****,v3.9.19.8123") switch: TRAPMGR: Link Down: 0/6

Jan 16 05:58:57 192.168.***.*** ("US8,f09fc2c6****,v3.9.19.8123") switch: DOT1S: Port (6) inst(0) role changing from ROLE_DESIGNATED to ROLE_DISABLED

Jan 16 05:59:00 192.168.***.*** ("US8,f09fc2c6****,v3.9.19.8123") switch: TRAPMGR: Link Up: 0/6

Jan 16 05:59:00 192.168.***.*** ("US8,f09fc2c6****,v3.9.19.8123") switch: DOT1S: Port (6) inst(0) role changing from ROLE_DISABLED to ROLE_DESIGNATED

Jan 16 05:59:26 192.168.***.*** ("US8,f09fc2c6****,v3.9.19.8123") switch: DOT1S: Port (6) inst(0) role changing from ROLE_DESIGNATED to ROLE_DISABLED

Jan 16 05:59:26 192.168.***.*** ("US8,f09fc2c6****,v3.9.19.8123") switch: TRAPMGR: Link Down: 0/6

Jan 16 05:59:27 192.168.***.*** ("US8,f09fc2c6****,v3.9.19.8123") switch: TRAPMGR: Link Up: 0/6

Jan 16 05:59:27 192.168.***.*** ("US8,f09fc2c6****,v3.9.19.8123") switch: DOT1S: Port (6) inst(0) role changing from ROLE_DISABLED to ROLE_DESIGNATED

Jan 16 06:00:07 192.168.***.*** ("US8,f09fc2c6****,v3.9.19.8123") switch: DOT1S: Port (6) inst(0) role changing from ROLE_DESIGNATED to ROLE_DISABLED

Jan 16 06:00:07 192.168.***.*** ("US8,f09fc2c6****,v3.9.19.8123") switch: TRAPMGR: Link Down: 0/6

Jan 16 06:00:10 192.168.***.*** ("US8,f09fc2c6****,v3.9.19.8123") switch: TRAPMGR: Link Up: 0/6

Jan 16 06:00:10 192.168.***.*** ("US8,f09fc2c6****,v3.9.19.8123") switch: DOT1S: Port (6) inst(0) role changing from ROLE_DISABLED to ROLE_DESIGNATE

Reply

Answer 10

Jan 17, 2018 7:57 AM in response to staengfoenster

Hate to do an "us too" post but we are experiencing this issue with the 4K units when plugged into Ethernet. We have always used Wifi in the past but on these 7 new 4K units we purchased, we decided to go over Ethernet and the network guys are not terribly happy. I currently have an open case with AppleCare Enterprise. I have tried putting two of them on the beta of tvOS 11.2.5 and that hasn't been helpful. I have also tried using a static IP on one of them per AppleCare Enterprise's request and that hasn't worked either. I am going to work with them some more today hopefully. Here's our switch logs if that helps anyone:

Trap logging: level informational, 124860 message lines logged

Logging to 10.xxx.xxx.xxx (udp port 514, audit disabled,

link up),

124813 message lines logged,

0 message lines rate-limited,

0 message lines dropped-by-MD,

xml disabled, sequence number disabled

filtering disabled

Logging Source-Interface: VRF Name:

Log Buffer (4096 bytes):

ace GigabitEthernet5/0/18, changed state to down

Jan 17 15:46:35.030: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet3/0/24, changed state to up

Jan 17 15:46:35.544: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/33, changed state to up

Jan 17 15:46:35.942: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/16, changed state to down

Jan 17 15:46:37.864: %LINK-3-UPDOWN: Interface GigabitEthernet5/0/18, changed state to up

Jan 17 15:46:37.945: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/16, changed state to up

Jan 17 15:46:38.868: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/18, changed state to up

Jan 17 15:46:42.982: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/16, changed state to down

Jan 17 15:46:43.985: %LINK-3-UPDOWN: Interface GigabitEthernet5/0/16, changed state to down

Jan 17 15:46:47.823: %LINK-3-UPDOWN: Interface GigabitEthernet5/0/16, changed state to up

Jan 17 15:46:48.826: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/16, changed state to up

Jan 17 15:46:55.342: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/18, changed state to down

Jan 17 15:46:57.348: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/18, changed state to up

Jan 17 15:47:06.387: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/16, changed state to down

Jan 17 15:47:08.387: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/16, changed state to up

Jan 17 15:47:20.935: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/33, changed state to down

Jan 17 15:47:20.946: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet3/0/24, changed state to down

Jan 17 15:47:21.949: %LINK-3-UPDOWN: Interface GigabitEthernet5/0/33, changed state to down

Jan 17 15:47:21.949: %LINK-3-UPDOWN: Interface GigabitEthernet3/0/24, changed state to down

Jan 17 15:47:25.189: %LINK-3-UPDOWN: Interface GigabitEthernet3/0/24, changed state to up

Jan 17 15:47:25.745: %LINK-3-UPDOWN: Interface GigabitEthernet5/0/33, changed state to up

Jan 17 15:47:26.193: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet3/0/24, changed state to up

Jan 17 15:47:26.748: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/33, changed state to up

Jan 17 15:47:43.069: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/18, changed state to down

Jan 17 15:47:43.069: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet3/0/24, changed state to down

Jan 17 15:47:43.187: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/33, changed state to down

Jan 17 15:47:44.068: %LINK-3-UPDOWN: Interface GigabitEthernet5/0/18, changed state to down

Jan 17 15:47:45.071: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet3/0/24, changed state to up

Jan 17 15:47:45.187: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/33, changed state to up

Jan 17 15:47:47.906: %LINK-3-UPDOWN: Interface GigabitEthernet5/0/18, changed state to up

Jan 17 15:47:48.906: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/18, changed state to up

Jan 17 15:47:52.985: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/16, changed state to down

Jan 17 15:47:53.985: %LINK-3-UPDOWN: Interface GigabitEthernet5/0/16, changed state to down

Jan 17 15:47:57.788: %LINK-3-UPDOWN: Interface GigabitEthernet5/0/16, changed state to up

Jan 17 15:47:58.788: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/16, changed state to up

Jan 17 15:48:05.548: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/18, changed state to down

Jan 17 15:48:07.551: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/18, changed state to up

Jan 17 15:48:16.628: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/16, changed state to down

Jan 17 15:48:18.631: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet5/0/16, changed state to up

Reply

Answer 11

nickguy

Level 1

4 points

Jan 31, 2018 6:18 PM in response to blackholemac

Same behavior with a 4th gen ATV. Port is flapping several times /minute same issue on multiple Extreme switches. Only a 100 meg port. Duplex/speed/hard coding of mac addresses/ all fail. Wireless in my case is probably faster than the Ethernet connection but it is just annoying. Please fix.

Reply

Answer 12

nickguy

Level 1

4 points

Jan 31, 2018 6:20 PM in response to nickguy

More logs

02/01/2018 02:06:18.61 <Info:vlan.msgs.portLinkStateUp> Port 4 link UP at speed 100 Mbps and full-duplex

02/01/2018 02:06:16.42 <Info:vlan.msgs.portLinkStateDown> Port 4 link down

02/01/2018 02:05:30.27 <Info:vlan.msgs.portLinkStateUp> Port 4 link UP at speed 100 Mbps and full-duplex

02/01/2018 02:05:28.41 <Info:vlan.msgs.portLinkStateDown> Port 4 link down

02/01/2018 02:05:03.13 <Info:vlan.msgs.portLinkStateUp> Port 4 link UP at speed 100 Mbps and full-duplex

02/01/2018 02:05:00.92 <Info:vlan.msgs.portLinkStateDown> Port 4 link down

02/01/2018 02:04:18.90 <Info:vlan.msgs.portLinkStateUp> Port 4 link UP at speed 100 Mbps and full-duplex

02/01/2018 02:04:17.05 <Info:vlan.msgs.portLinkStateDown> Port 4 link down

02/01/2018 02:03:52.21 <Info:vlan.msgs.portLinkStateUp> Port 4 link UP at speed 100 Mbps and full-duplex

02/01/2018 02:03:50.01 <Info:vlan.msgs.portLinkStateDown> Port 4 link down

02/01/2018 02:03:03.78 <Info:vlan.msgs.portLinkStateUp> Port 4 link UP at speed 100 Mbps and full-duplex

02/01/2018 02:03:01.94 <Info:vlan.msgs.portLinkStateDown> Port 4 link down

02/01/2018 02:02:36.94 <Info:vlan.msgs.portLinkStateUp> Port 4 link UP at speed 100 Mbps and full-duplex

02/01/2018 02:02:34.74 <Info:vlan.msgs.portLinkStateDown> Port 4 link down

02/01/2018 02:01:48.58 <Info:vlan.msgs.portLinkStateUp> Port 4 link UP at speed 100 Mbps and full-duplex

02/01/2018 02:01:46.74 <Info:vlan.msgs.portLinkStateDown> Port 4 link down

02/01/2018 02:01:22.16 <Info:vlan.msgs.portLinkStateUp> Port 4 link UP at speed 100 Mbps and full-duplex

02/01/2018 02:01:19.95 <Info:vlan.msgs.portLinkStateDown> Port 4 link down

02/01/2018 02:00:33.71 <Info:vlan.msgs.portLinkStateUp> Port 4 link UP at speed 100 Mbps and full-duplex

02/01/2018 02:00:31.86 <Info:vlan.msgs.portLinkStateDown> Port 4 link down

02/01/2018 02:00:07.41 <Info:vlan.msgs.portLinkStateUp> Port 4 link UP at speed 100 Mbps and full-duplex

02/01/2018 02:00:05.15 <Info:vlan.msgs.portLinkStateDown> Port 4 link down

02/01/2018 01:59:43.59 <Info:vlan.msgs.portLinkStateUp> Port 4 link UP at speed 100 Mbps and full-duplex

02/01/2018 01:59:41.75 <Info:vlan.msgs.portLinkStateDown> Port 4 link down

02/01/2018 01:59:16.65 <Info:vlan.msgs.portLinkStateUp> Port 4 link UP at speed 100 Mbps and full-duplex

02/01/2018 01:59:14.44 <Info:vlan.msgs.portLinkStateDown> Port 4 link down

Reply

Answer 13

robmille

Level 1

9 points

Mar 1, 2018 8:48 PM in response to Andypeppa

I have this same log spamming behavior going on with 2 Apple TVs, one on a Cisco 3650-X switch that has PoE, and one connected to a non-PoE Cisco 2960S switch. I've played around with disabling power settings. Setting speed & duplex etc. No impact whatsoever.

Apple TV perpetually enables it's NIC, put's it's MAC address on the switch (presumably checking it's DHCP lease, and probably doing some bonjour or air play checks etc), and then shuts the NIC off shortly after.

I filed a Radar with Apple as a bug. There's no excuse for this network noise.

Reply

Answer 14

Mar 7, 2018 5:08 AM in response to staengfoenster

I wish I had better to report...AppleCare Enterprise wanted me to give the latest tvOS 11.3 beta 15L5198d a whirl saying that things may have improved. I attempted it and all seemed to work well on a single test unit here for one day. I then extended the test out to our intermediate school and left for a night. It was still flapping in the wind. Checked back on my single test unit and now it's doing the same. Back to the drawing board. This case seems to be moving at a glacial pace so others please escalate as well. The good news is, the rep seemed to reproduce the issue and is also working on it from his end.

By the way, I don't post on here very frequently and when I wrote this, I realized I logged in with a different Apple ID. I'm 'blackholemac'

Reply

Answer 15

GpaJake

Level 1

4 points

Mar 7, 2018 6:44 AM in response to Andypeppa

My Google fiber router is very limited in what I can see but I also have noticed that the AppleTV 4K switches back and forth from 100Mbps to 1Gbps while in idle mode. At the same time I lose connection to the internet even though the router still claims an IP address for the ATV. Sometimes it works to restart the unit. Sometimes I have to unplug the unit from the router then plug it back in. This problem did not occur until I recently switched to 1Gbps internet. Previously it was connected to an Airport Extreme running off 75Mbps cable internet. It is very frustrating and I wish I could find a way to keep this from happening.

Reply