HOA fee run ISP problem..

[fyun89]

Hi, I play internet game during my precious after-work free time and I've been getting this "ping spike" recently. This problem started when we received complimentary speed increase from the ISP named Greenfield communication.

I have been following up with Greenfield's tech support and a cable guy came to see the problem as well. However they did not find any problem after SEVERAL repeated troubleshoot procedures.

I even contacted Riot, the company behind the game I play, and showed them results of Tracert and ping tests. They said that the problem seems to occur at my ISP's end as the hop within my ISP's network show high, intermittent ping.

I'm not a technical expert, but I am a bit technically inclined person. The following reasons are the reason for my suspect that it is the Greenfield's problem.
1. This happened right after the speed increase
2. Tracert + ping test (MTR used) shows intermittent ping spike at Greenfield's network
3. My friend who lives right across the street (not Greenfield, ATT) in another neighborhood is not having the lag when I get it.
4. Riot tech support says this is my ISP's problem after reviewing their proprietary test software that I ran for them.

The reason why Greenfield is making me upset is:
1. They are keep making me do the same troubleshoot over and over again (reset router, restart computer, try another computer, etc etc)
2. I did what they told me to do (troublshoot procedures and cable guy visit) and still claiming my home's network problem although they found no problem. In other words, not admitting their fault.
3. Although Im not entirely sure, but they claimed that one of the homes they were serving was under DDOS attack during my latest disconnection. Less than fifteen minutes later they solved the problem... I mean I've never seen any company overcoming DDOS attack in few minutes like that... Unless they were attacking me.. But there is a bit or a stretch there.. Im just really angry and i cant trust them anymore.
4. Riot's explanation were perfectly understandable while Greenfield's explations all sounded BS. I tried to be unbiased but Greenfield's statements all sounded like excuses. One example: Riot said the "worst" column of the MTR application showed highest ping rate (~300) and we could assume that that be the cause of the intermittent ping spike/disconnection. However Greenfield says to look at the Best ping result... Which I completely didnt understand why you would look at Best ping rate when we're looking for intermittent lag...


The problem is that we cannot switch ISP. It would not be a wise choice as our internet fee is combined in our HOA fee. If we get another ISP, we'd be paying two ISPs at once...

Im just really angry and I even filed FTC complaint but I'm not sure if they would take action. Does anyone have any advice on how to tackle this problem? Or is anyone who live in community where they use Greenfield internet and have this same problem?

Man Im so angry.. I feel like they're taking advantage of the "lock-in" contract with our comminity and know that whether their service is terrible we the residents cant do anything..

 

[Shaun o]

The problem is most people do not see is, when a ISP provides a free speed increase to it`s customers.
It involves having to upgrade what we call the Mid hubs with new rack or circuit board hardware.

The Hub, or Mid hub.

Is the point where where a set area, and number of fiber optic connections meet say example for 500 fiber optic connections from street 1,2,4,5.

On the Mid hub it`s self the traffic load or balance including the new hardware racks must be configured correctly.
As this point can can cause a bandwidth problem, or result in a high latency or ping at peak times where usage is high.

The way to test this from a user point is to pick a time where most other users are asleep ect, or out of peak times.

If you find that the connection is better and the latency or ping is reduced or lower than peak times.
It suggest that the ISP you are signed up with needs to take a look at the Mid hub if your own connection has been tested and no problem by there visits reveal nothing.

Have you got a trace route of an IP address of a server you use for gaming.
To see based on the amount of hops where the high latency is happening fyun89.

It should tell you and confirm if it is to do with the ISP network or the point where it leaves it to cross to another network.

If you could can you post one ?

Some cable modem routers include the facility to do a trace route if you log into the router as part of diagnosis.

Also check the status if the router provides it of your channels that are bound, the voltage each one is reporting and the amount of line noise in DBv.

As said found by logging into the router modem if you have one.

 

[smorizio]

with cable modems make sure your modem is on the first spitter in your home. make sure the splitter is a newer unit that wont be blocking the cable modem signal. also check or have your cable guy check that the cable to your home is on tight and not rusted or wire if it above ground chewed up. also try using open dns or gooogle dns to see if the error stops. also most of the newer modems have to be used. when comcast and other give you a speed bump there now sending more data over more bonded channels. older modems only have two to four channels for bonding the newer ones have 8 and now there ones out that have 16 channels. they also have to have the newest firmware on them. make sure the isp pushed the right firmware for the modem in your home. for your hoa and isp issue. look at see if comcast or verizion is cheaper and or better. sometime now doing disk for tv and isp for cable can be better and cheaper. then go around have people sign a partion to switch. if you have the siguture the hoa will have to change isp.

 

[fyun89]

The problem is most people do not see is, when a ISP provides a free speed increase to it`s customers.
It involves having to upgrade what we call the Mid hubs with new rack or circuit board hardware.

The Hub, or Mid hub.

Is the point where where a set area, and number of fiber optic connections meet say example for 500 fiber optic connections from street 1,2,4,5.

On the Mid hub it`s self the traffic load or balance including the new hardware racks must be configured correctly.
As this point can can cause a bandwidth problem, or result in a high latency or ping at peak times where usage is high.

The way to test this from a user point is to pick a time where most other users are asleep ect, or out of peak times.

If you find that the connection is better and the latency or ping is reduced or lower than peak times.
It suggest that the ISP you are signed up with needs to take a look at the Mid hub if your own connection has been tested and no problem by there visits reveal nothing.

Have you got a trace route of an IP address of a server you use for gaming.
To see based on the amount of hops where the high latency is happening fyun89.

It should tell you and confirm if it is to do with the ISP network or the point where it leaves it to cross to another network.

If you could can you post one ?

Some cable modem routers include the facility to do a trace route if you log into the router as part of diagnosis.

Also check the status if the router provides it of your channels that are bound, the voltage each one is reporting and the amount of line noise in DBv.

As said found by logging into the router modem if you have one.

Sorry about the late reply. I was really busy past few weeks I couldn't really answer your question and give you additional info.

Yes. I'm sorry that the post I wrote is very long. But I have indicated that I have performed traceroute and ping test (using WinMTR).

Here you will find that "border1.te4-4.greencomm-2.lax008.pnap.net" having highest WRST ping. Also, I asked them if anything went wrong with their system and they said their side is having NO PROBLEM. This is a small company and I am always connected to same Technical Service guy. So I assume that giving him solution to this problem would change anything. I lost my temper last time and didn't end the call in nice fashion.. I don't want to call them back again.

Is there any legal method other than reporting to FTC to fix this matter?

one of the MTR test I have is this:
|------------------------------------------------------------------------------------------|

| WinMTR statistics |

| Host - % | Sent | Recv | Best | Avrg | Wrst | Last |

|------------------------------------------------|------|------|------|------|------|------|

| Linksys25939 - 0 | 74 | 74 | 0 | 1 | 3 | 1 |

| dynamic-112-1.ful.ca.us.egreenfield.com - 0 | 74 | 74 | 1 | 2 | 11 | 2 |

| 108.174.208.57 - 0 | 74 | 74 | 9 | 11 | 15 | 10 |

|border1.te4-4.greencomm-2.lax008.pnap.net - 0 | 74 | 74 | 10 | 19 | 219 | 10 |

| core2.po2-20g-bbnet2.lax015.pnap.net - 0 | 74 | 74 | 10 | 12 | 14 | 11 |

| xe-2-2-0.er4.lax112.us.above.net - 0 | 74 | 74 | 10 | 12 | 18 | 13 |

| ae10.cr2.lax112.us.zip.zayo.com - 0 | 74 | 74 | 11 | 13 | 31 | 12 |

| ae16.er2.lax112.us.zip.zayo.com - 0 | 74 | 74 | 11 | 14 | 64 | 12 |

| 64.125.170.174.t01598-01.above.net - 0 | 74 | 74 | 11 | 14 | 36 | 13 |

| 104.160.133.35 - 0 | 74 | 74 | 20 | 25 | 46 | 23 |

| ae36-br01.chi01.riotdirect.net - 0 | 74 | 74 | 65 | 67 | 74 | 70 |

| 104.160.131.1 - 0 | 74 | 74 | 64 | 68 | 77 | 64 |

|________________________________________________|______|______|______|______|______|______|

WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider



Here's another data from this VPN program called WTFast. (in order: packetloss, Sent, Received, Best, Avg, Last, Spikes, Flux)

dynamic-112-1.ful.ca.us.egreenfield.com 2 0.48 1470 1463 1 9 300 12 2 200 9.46
108.174.208.57 3 0.07 1470 1469 9 17 612 25 3 389 19.73
border1.te4-4.greencomm-2.lax008.pnap.net 4 0.20 1470 1467 10 27 542 124 144 109 36.53
core1.be1-40g-bbnet1.lax010.pnap.net 5 0.14 1470 1468 11 19 403 26 2 355 13.48
be4212.ccr41.lax04.atlas.cogentco.com 6 0.00 1470 1470 11 19 152 15 2 132 6.80
be2964.ccr21.lax01.atlas.cogentco.com 7 0.07 1470 1469 11 19 47 24 1 29 5.12
be2180.ccr23.lax05.atlas.cogentco.com 8 0.00 1470 1470 11 19 40 13 0 0 5.04
38.104.84.254 9 0.07 1470 1469 11 19 53 19 0 0 5.14
Target WTFast.com 10 0.00 1470 1470 11 18 54 28 0 0 5.50

UPDATE: here's another one that I've done for a bit longer time
|------------------------------------------------------------------------------------------|

| WinMTR statistics |

| Host - % | Sent | Recv | Best | Avrg | Wrst | Last |

|------------------------------------------------|------|------|------|------|------|------|

| LinksysY - 0 | 678 | 678 | 0 | 2 | 18 | 8 |

| dynamic-112-1.ful.ca.us.egreenfield.com - 1 | 674 | 673 | 1 | 3 | 206 | 3 |

| 108.174.208.57 - 1 | 670 | 668 | 9 | 11 | 31 | 16 |

|border1.te4-4.greencomm-2.lax008.pnap.net - 1 | 674 | 673 | 9 | 21 | 379 | 15 |

| core1.po2-20g-bbnet2.lax012.pnap.net - 1 | 674 | 673 | 10 | 13 | 330 | 12 |

| ae-19.r01.lsanca07.us.bb.gin.ntt.net - 1 | 674 | 673 | 10 | 14 | 423 | 16 |

|ix-ae-5-0.tcore2.LVW-Los-Angeles.as6453.net - 1 | 670 | 668 | 10 | 17 | 88 | 21 |

| 74.125.50.122 - 1 | 674 | 673 | 21 | 25 | 219 | 25 |

| 209.85.245.191 - 1 | 674 | 673 | 21 | 25 | 220 | 28 |

| 216.58.215.139 - 1 | 674 | 673 | 21 | 25 | 447 | 22 |

| google-public-dns-a.google.com - 1 | 674 | 673 | 21 | 27 | 948 | 23 |

|________________________________________________|______|______|______|______|______|______|

WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider

 

[bill001g]

Pretty much your data proves you have no issues....at least at the time you were testing. The key numbers to watch are packet loss which you have maybe 1 and the variation in best and average times. Yours shows you have very little deviation between these numbers. To have issues you would see many 100s of ms difference.

You also have to be very careful about interpreting data from intermediate hops. If say hop 3 has a problem and is dropping 10% of all the data you would see at least 10 % data loss in hop 4 to the end. If you see a problem in only one hop and not in further hops just means that particular device is delay or dropping response to ping. It is very common for a device to not process a ping when it has more important tasks to do...like passing actual traffic for users.

Your problem is you need to catch this when the problem occurs and many times by the time you see it and start a problem it will have gone.

I would do this function manually. Open a bunch of cmd windows and leave a continuous ping running to a number of hops in the trace. When you see the problem in the game quickly stop all the windows and see if anything significant shows up.


A key thing to remember with a game is it is not actually running ping in most cases. It is measure the delays it sees between packets. Since this is done by the game client itself it include client and server delays as well as network delays. Many times these can show issues but the network does not actually have any problem. I have see crazy stuff fix this, you will find post where people adjust the graphics settings to fix "ping" problems in a game. It just shows that the game may claim a network problem when it is not.

 

[fyun89]

The fact that this had suddenly happened right after the speed upgrade, WinMTR (Ping+traceroute) showing such data, and verifying with other technician (the game's tech support) pretty much clearly points at my ISP.

This ping spike seem to happen even more severly at high-usage time (6PM - 9 PM). When I'm doing Skype or Facetime on my PHONE when this ping spike happesn, the call gets disconnected too.

I can attest this problem is on ISP. However, my ISP won't do anything for me. Is there a legal step that I can take in this case?

Graphics problem is not even a problem. My other computers show same ping spike (and other application such as Skype and Facetime disconnects too)

 

[bill001g]

You need to get something you can prove it there fault or you will get you saying it is there problem and them saying it is your router or pc and you get nowhere.

You say you have a winMTR that clearly shows it is the ISP but what you posted show no issues at all. You need one that actually show the problem start in the ISP node and causes issues all the way to the end of the trace.

 

[fyun89]

The WinMTR does show the hops within their network with record of intermittent ping spike in the column WORST (ones with word"green" in the address). And I think the suggestion you gave me before (opening bunch of CMD and ping test all the hops separately) is what this winMTR does; but in a more ergonomic fashion...

How is MTR any different from running ping tests separately on each hop? Is there another tool that I can use to create more evidence of this problem?

Here's another MTR data I just got
|------------------------------------------------------------------------------------------|

| WinMTR statistics |

| Host - % | Sent | Recv | Best | Avrg | Wrst | Last |

|------------------------------------------------|------|------|------|------|------|------|

| LinksysY - 0 | 192 | 192 | 0 | 3 | 17 | 2 |

| dynamic-112-1.ful.ca.us.egreenfield.com - 1 | 188 | 187 | 1 | 5 | 28 | 3 |

| 108.174.208.57 - 1 | 188 | 187 | 9 | 14 | 612 | 11 |

|border1.te4-4.greencomm-2.lax008.pnap.net - 1 | 188 | 187 | 9 | 23 | 542 | 201 |

| core2.po2-20g-bbnet2.lax012.pnap.net - 1 | 188 | 187 | 11 | 13 | 38 | 14 |

| 12.91.226.17 - 1 | 188 | 187 | 10 | 18 | 75 | 44 |

| cr2.la2ca.ip.att.net - 1 | 188 | 187 | 12 | 17 | 41 | 14 |

| 12.123.132.229 - 1 | 188 | 187 | 11 | 38 | 172 | 12 |

| 12.252.12.22 - 1 | 188 | 187 | 21 | 23 | 53 | 23 |

| 216.239.59.217 - 1 | 188 | 187 | 21 | 24 | 50 | 24 |

| 72.14.233.225 - 1 | 188 | 187 | 21 | 23 | 48 | 24 |

| google-public-dns-a.google.com - 1 | 188 | 187 | 21 | 23 | 47 | 23 |

|________________________________________________|______|______|______|______|______|______|

WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider


Also, I found another net analysis tool called ICSI Netalyzr
The result:
•Certain TCP protocols are blocked in outbound traffic
•The network indicated bursts of packet loss
•Your computer's clock is slightly fast

Detail on the second bullet:
"During most of Netalyzr's execution, the client continuously measures the state of the network in the background, looking for short outages. During testing, the client observed 2 such outages. The longest outage lasted for 2.8 seconds. This suggests a general problem with the network where connectivity is intermittent. This loss might also cause some of Netalyzr's other tests to produce incorrect results. "

 

[bill001g]

It appears you are still misreading the trace.

The last hop show no problems at all. That means the traffic is getting to the end device with no issues. When this happens it really does not matter if the intermediate hops show problems.

You must show a problem that really is affecting you and that is only traffic loss to the end nodes.

Your other tool is still pretty much worthless. It yells I see network problem but does not show what it is. Your ISP will say it is your modem/router etc so it is of little use.