Internet disconnecting multiple times an hour for the past week

okay but like... does anyone know WHY this happened?!

like legit i'm happy it's back to normal working ability .. but there's a reason and for me missing a whole month of my service.. I would really like to know why.

Still some problems with latency 71.147.188.41:

|------------------------------------------------------------------------------------------|
|                                      WinMTR statistics                                   |
|                       Host              -   %  | Sent | Recv | Best | Avrg | Wrst | Last |
|------------------------------------------------|------|------|------|------|------|------|
|                  dsldevice.attlocal.net -    0 | 2898 | 2898 |    0 |    0 |   15 |    0 |
|76-225-148-1.lightspeed.irvnca.sbcglobal.net -    0 | 2899 | 2899 |   18 |   21 |   50 |   23 |
|                           71.147.188.41 -    0 | 2893 | 2893 |   18 |   22 | 4299 |   19 |
|                           12.242.115.33 -    0 | 2898 | 2898 |   21 |   21 |   42 |   22 |
|                            12.245.64.10 -    0 | 2899 | 2899 |   21 |   26 |  155 |   22 |
|              ae1-br02-eqla1.as57976.net -    0 | 2892 | 2892 |   21 |   36 | 3958 |   22 |
|                           137.221.68.87 -    0 | 2898 | 2898 |   21 |   24 |  147 |   21 |
|        lax-eqla1-ia-bons-02.as57976.net -    0 | 2899 | 2899 |   22 |   22 |   35 |   22 |
|                           24.105.30.129 -    0 | 2899 | 2899 |   21 |   22 |   34 |   23 |
|________________________________________________|______|______|______|______|______|______|
   WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider

Still some problems with latency 71.147.188.41:

...

71.147.188.41 -    0 | 2893 | 2893 |   18 |   22 | 4299 |   19 |

12.242.115.33 -    0 | 2898 | 2898 |   21 |   21 |   42 |   22 |

Where? I don't see any latency issues.   It has an average of 22 and the next hop has a max of 42.  This is not an indicator of anything at all.  Pointing at this sort of thing as an issue is a quick way to get your queries ignored.

Hi JefferMC,

Thank you for your interest and reviewing this matter. Maybe I am not reading the report the same way.

|------------------------------------------------|------|------|------|------|------|------|
|                  dsldevice.attlocal.net -    0 | 2898 | 2898 |    0 |    0 |   15 |    0 |
|76-225-148-1.lightspeed.irvnca.sbcglobal.net -    0 | 2899 | 2899 |   18 |   21 |   50 |   23 |
|                           71.147.188.41 -    0 | 2893 | 2893 |   18 |   22 | 4299 |   19 |
|                           12.242.115.33 -    0 | 2898 | 2898 |   21 |   21 |   42 |   22 |

4299 to me seems high for max latency. I get that the next hop is fine, but that is not the IP address that I was mentioning. If I am misunderstanding something, let's use this as an opportunity to educate the forum. I know that after this past experience, I would like to learn more to help more quickly identify, communicate, and resolve the issue.

Also, when you stated, "Pointing at this sort of thing as an issue is a quick way to get your queries ignored." is neither constructive nor helpful when it was myself that provided the data that led the issue getting attention it needed. 

... it was myself that provided the data that led the issue getting attention it needed. 

Puhlease.  You were number 8 or 9 in the list of people having an issue.  What made me include your data as significant was (1) that there were so many people ahead of you and (2) the consistent missing responses all the way out to the last hop (recv < sent consistently).  Thank you for the data, it was helpful.  But it was just one piece of a puzzle being assembled well before you posted.  And, I actually still have no idea if AT&T actually did anything as a result of this thread or if it just cured itself, or it was fixed without reference to our conversation here.

let's use this as an opportunity to educate the forum.

Fair enough.  It bears repeating.

Recognize that routers are busy bees and have many things to do.  As such they prioritize their work.  And one of the things that is low on the priority scale is ICMP traffic directed at them.  A thing even lower is an ICMP packet with an error condition.

Traceroute works by sending out ICMP packets that are designed to error out after a designated number of hops.  When a hop gets a "TTL expired" packet, it may decide to generate an error message, if it has time and hasn't hit its limit of such responses during the current time interval.  If it decides to drop the packet, you won't get a response.  PING uses ICMP, but doesn't play with the TTL count to cause the packets to expire.  Winmtr uses a combination of the two.

The first thing to remember about these displays is that if a hop truly has an issue, that issue will continue to manifest itself for each and every hop behind it.  I cannot stress this foundational point enough:  If hop 3 truly has an issue, then traffic destined for hop 4 via hop 3, will be affected by that issue.  Similarly, traffic destined for port 9 will also pass through hop 3 (and 4 through 8). 

So, looking back at the WinMTR you posted as evidence that 71.147.188.41 "still has some latency problems", you will note that max latency for every hop past it are: 42, 155, 36, 147, 22 and 34.  So, the last node, the farthest one away, has a max latency of 34 ms.  Meaning that the packet went through 3, got to 9, came back to 3, and got to you in 34 ms.  So, the true max latency for hop 3 is 34 ms or less.  (Actually, it's 22 or less, since hop 8 never exceeded 22).

Also note that it only takes one packet to make max latency go high.  You really shouldn't draw any conclusion on one data point. 

I have been pretty busy this past few days so I haven't had time to see if the issue has persisted, but I can confirm that on Saturday the 5th night I had stable internet during a game session from 7-9pm. I had a single lag spike but it wasn't enough to disconnect me. Seeing everyone else's posts about the situation improving gives me hope.

Aren't you just a bit late to this party Qudus?

It appears the problem may be back again see below:

Anyone else seeing this problem? 76-225-148-1.lightspeed.irvnca.sbcglobal.net

|------------------------------------------------------------------------------------------|
|                                      WinMTR statistics                                   |
|                       Host              -   %  | Sent | Recv | Best | Avrg | Wrst | Last |
|------------------------------------------------|------|------|------|------|------|------|
|                  dsldevice.attlocal.net -    0 |  243 |  243 |    0 |    0 |   14 |    0 |
|76-225-148-1.lightspeed.irvnca.sbcglobal.net -    7 |  197 |  185 |   18 |   28 |   52 |   22 |
|                           71.147.188.41 -    7 |  192 |  179 |   18 |   30 |  625 |   19 |
|                           12.242.115.33 -   13 |  165 |  145 |   21 |   42 |  122 |   21 |
|                            12.245.64.10 -   18 |  145 |  120 |   21 |   47 |  172 |   22 |
|              ae1-br02-eqla1.as57976.net -   11 |  173 |  155 |   21 |   53 |  166 |   46 |
|                           137.221.68.87 -   15 |  157 |  135 |   21 |   48 |  135 |   22 |
|        lax-eqla1-ia-bons-02.as57976.net -   15 |  157 |  135 |   22 |   40 |  122 |   23 |
|                           24.105.30.129 -   16 |  153 |  130 |   21 |   40 |  122 |   22 |
|________________________________________________|______|______|______|______|______|______|
   WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider

Something is going on again today. See the report below:

|------------------------------------------------------------------------------------------|
|                                      WinMTR statistics                                   |
|                       Host              -                         %  | Sent | Recv | Best | Avrg | Wrst | Last |
|----------------------------------------       --------|------|------|------|------|------|------|
|                  dsldevice.attlocal.net -                    0 |   79 |   79 |    0 |    0 |    3 |    0 |
|76-225-148-1.lightspeed.irvnca.sbcglobal.net -   15 |   49 |   42 |   19 |   24 |   39 |   21 |
|                           71.147.188.41 -                     11 |   56 |   50 |   18 |   25 |   38 |   38 |
|                           12.242.115.33 -                     31 |   36 |   25 |   21 |   46 |  131 |   26 |
|                            12.245.64.10 -                      18 |   46 |   38 |   22 |   53 |  127 |  127 |
|              ae1-br02-eqla1.as57976.net -           19 |   44 |   36 |   21 |   56 |  151 |   32 |
|                           137.221.68.87 -                     28 |   36 |   26 |   21 |   61 |  135 |   46 |
|        lax-eqla1-ia-bons-02.as57976.net -        28 |   36 |   26 |   22 |   38 |  100 |   47 |
|                           24.105.30.129 -                     19 |   44 |   36 |   21 |   47 |  130 |   32 |
|_______________________________________________|______|______|______|______|______|______|
   WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider

This is happening to me too on all devices and whether or not I am connected via Ethernet or Wifi.