For a year before March 2016 my connection on Infinity 1 with a HH5 type A was rock solid, virtually never dropping. But since March 2016 it drops around once per day, usually evenings but could be any time. Sometimes it'll go a few days, sometimes it'll drop several times in one day.
The HH5 light goes orange. The problem is, it never reconnects, unless I either power down the HH5 for a minute or disconnect the wall socket for a minute. Either method seems to clear the issue.
I tried my previous HH4 as a test, and that also suffered the same dropouts. So it's probably not the HH.
All the other comments on here seem to indicate the HH5 is able to reconnect by itself after a few minutes. But mine never does. I can see it trying, the light goes blue briefly, but always orange again until I break the link as mentioned.
Is this a clue as to what might technically be going wrong with with the connection?
Any ideas for how I can diagnose this? The quiet line test sounds silent to me. Here are the line speeds:
This is interesting.
I'm trying to sort out my dad's broadband connection, and he's using a BT Home Hub 5A too.
It seems unstable, and has dropped down to exactly the same IP profile as yours.
I'm tempted to dig out my HG612 and use that to connect instead of the HH 5A.
I don't think the profile has anything to do with losing connection.
It just dropped out, here are my stats just before the dropout, if that helps. I removed all Mac addresses.
16:28:34, 17 Feb. (60804.680000) Admin login successful by 192.168.1.75 on HTTP 16:28:29, 17 Feb. (60799.570000) CWMP: session closed due to error: No response 16:28:29, 17 Feb. (60799.530000) CWMP: Server URL: https://pbthdm.bt.mo; Connecting as user: ACS username 16:28:29, 17 Feb. (60799.520000) CWMP: Session start now. Event code(s): '4 VALUE CHANGE' 16:28:07, 17 Feb. (60778.030000) New GUI session from IP 192.168.1.75 16:27:58, 17 Feb. (60768.800000) CWMP: session closed due to error: No response 16:27:58, 17 Feb. (60768.700000) CWMP: Server URL: https://pbthdm.bt.mo; Connecting as user: ACS username 16:27:58, 17 Feb. (60768.690000) CWMP: Session start now. Event code(s): '4 VALUE CHANGE' 16:27:57, 17 Feb. (60767.970000) CWMP: Initializing transaction for event code 4 VALUE CHANGE 16:27:56, 17 Feb. (60766.670000) PPPoE is down after 1010 minutes uptime [Disconnected] 16:27:53, 17 Feb. (60764.130000) PPP LCP Send Termination Request [Peer not responding] 16:24:16, 17 Feb. (60546.600000) Device disconnected: Hostname: android-bfd1ec84b266122a IP: 192.168.1.202 MAC: xxx 16:24:16, 17 Feb. (60546.600000) Device disconnected: Hostname: android-8e0adec7a367df15 IP: 192.168.1.81 MAC: xxx 16:24:16, 17 Feb. (60546.590000) Device disconnected: Hostname: Chris-mobile IP: 192.168.1.200 MAC: xxx 16:24:16, 17 Feb. (60546.590000) Device disconnected: Hostname: Marcus_PC IP: 192.168.1.182 MAC: xxx 16:24:16, 17 Feb. (60546.590000) Device disconnected: Hostname: COM-MID1 IP: 192.168.1.184 MAC: xxx 16:24:16, 17 Feb. (60546.590000) Device disconnected: Hostname: Heatmiser-00-90-c2-e8-09-53 IP: 192.168.1.63 MAC:xxx 16:24:16, 17 Feb. (60546.590000) Device disconnected: Hostname: kindle-e9a8299d3 IP: 192.168.1.69 MAC: xxx 16:19:01, 17 Feb. IN: BLOCK  Remote administration (TCP [188.8.131.52]:41565->[184.108.40.206]:22 on ppp3)
The point I was making, was that are these Home Hubs dropping connection in a similar fashion, enough to persuade the DLM to drop down to 32.4Mb, which it believes is the correct speed for an unstable Home Hub? I could be wrong, but that was my line of thought.
I decided to connect the Huawei HG612 to the master socket instead, and the HH 5A connects to the HG612, to eliminate the 5A as a potential source of the problem. I'm going to monitor the stability carefully to see if this helps.
I don't think a third party router is an option as I get BT Sport and Sky Sport via multicast?
I have just a master socket (it's a fairly recent house), and the faceplate was supplied and fitted by Openreach.
I already tried my old HH4 which made no difference to the loss of connection and need for power cycling.
You're right, the DLM is supposed to slow the speed, to maintain the connection.
Which begs the questions, why the repeated loss of connection, and why do I need to power cycle the HH each time?
"If a line is unstable then DLM will take action to stabilise the line. "
So in my case it fails to do so? And why do I have to power cycle the hub to get it back?