Recently moved to new house, ping goes up at night

vensus

Distinguished
Apr 28, 2016
86
0
18,640
Hi there, I've recently moved over the last week to a new house and before setting anything up of my own.. There is the coax cable leading from the power line pole in front of our house, which leads to the side of our house and travels into the basement. This then connects into a 4 output splitter which has all the coax cables leading back out of the house and into each of our four bedrooms on the 2nd floor. These coax cables were installed within the last half year to a year ago from what the previous owner told us. They go into the walls and into a wall panel that has female to female adapter/connectors.

Now when i started setting things up here, I attached a 1 foot coax cable into the wall adapter which leads into our modem. Then a 1-2 foot ethernet cable into our router. I then had to drill into the corner of the flooring so one 35 feet ethernet cable drops right where our living room tv area is (used for swapping for tv/xbox/ps4) then a 50ft one leading into the basement and around a corner going back up into our sun room area for our mothers work computer. Then lastly 2 75ft cables leading into the ceiling to the attic, which leads to 2 of the 4 bedrooms up on our 2nd floor for two gaming computers.

Before I continue.. I've read that some ethernet cables should be a good foot or two away from any uv lights or other eletrical wiring, which I've done an okay job at doing in the basement and attic. But I was considering hanging the cables in the attic towards the roofing beams so they're far away from any other wires and if must be pull the basement one back into the living room and run it along the side of our carpet leading into the sun room area. I mention this because I've read other wires can cause interferce which might come into play with my issue?

Moving on.. I've started to notice that it would seem our internet is fine through the day with no loss of connection or anything that I can notice happening out of place. That is till the other night ago I noticed while playing a game online that my normal ping of 70-80 to the games server went up to 100-120 and seems to stay there from the time frame of 7pm-12pm or so.. I checked my discord program ping and that also had jumped from my avg of 40-50 to east coast, to being 80+ range. Same thing was happening when trying speedtest.com to connect to the comcast host located in Chicago which normally is 15-17 ping/179-180mbps down/23-24mbps up but during that time frame the ping was now 45-50+ and oddly down/up speeds didn't seem affected. But now tonight I've noticed our ping has jumped up again around the same time and still hasn't went back to normal as of right now (9:05pm). Doing a speed test again right now though shows ping at 53ms/151mbps down/23mbps up, so it looks like maybe our down speeds are also being affected as well?

I'm wondering if this sounds like something to do with interferce that I could go fix with the basement/attic cables or if this sounds like something more then that? We moved from a few miles out of town into a few blocks from the main center of our town, before the move we'd not have ping or speed issues even while two or more people were hardcore gaming and on voip, etc. If I do have to have a tech come out and check anything over though I'd love to know what to tell them in terms of stuff to check, go over, etc. for this sort of issue rather then the normal tech check that they'll do with checking the line health and speeds while ping/speed isn't having issues.. (comcast techs)

Below is a bit of info on hardware we have running for our setup, all of our hardware was bough brand new from this past summer, even the ethernet cables. Also a few speed tests and tracert tests below, if there's anything else I should run for tests before the timeframe and during the timeframe of issues let me know and I'll run them.

Modem: netgear CM600
router: netgear WNDR4500v3
Ethernet cables: »www.amazon.com/gp/produc ··· HJ&psc=1
(using 2 75ft ones for the two going into attic, 1 30-35ft one leading to living room tv area, 1 50ft one to basement back into sun room area)

speedtest (before issue timeframe, earlier today)
http://www.speedtest.net/result/6915305020.png

speedtest (during issue timeframe, tonight a few mins ago)
http://www.speedtest.net/my-result/6915976207

line quality test (earlier today, before issue timeframe)
http://www.dslreports.com/pingtest/82467c46742c/3459768

line quality test (tonight, during issue timeframe)
http://www.dslreports.com/pingtest/3575432a39c6/3459805

tracert test to google.com (during issue time frame)
Tracing route to google.com [172.217.1.46]
over a maximum of 30 hops:

1 1 ms 1 ms 1 ms 192.168.1.1
2 9 ms 8 ms 9 ms 96.120.26.221
3 9 ms 9 ms 8 ms te-0-7-0-19-sur01.mendota.il.chicago.comcast.net
[162.151.34.109]
4 49 ms 53 ms 56 ms te-17-10-cdn04.mendota.il.chicago.comcast.net [6
9.139.202.150]
5 48 ms 52 ms 51 ms te-0-1-0-4-7-ar01.elmhurst.il.chicago.comcast.ne
t [162.151.92.209]
6 58 ms 57 ms 53 ms be-33491-cr01.chicago.il.ibone.comcast.net [68.8
6.92.33]
7 58 ms 50 ms 51 ms be-10506-cr02.350ecermak.il.ibone.comcast.net [6
8.86.86.229]
8 41 ms 39 ms 38 ms hu-0-16-0-1-pe04.350ecermak.il.ibone.comcast.net
[68.86.86.38]
9 42 ms 42 ms 44 ms as15169-2-c.350ecermak.il.ibone.comcast.net [66.
208.233.142]
10 * * * Request timed out.
11 57 ms 53 ms 61 ms 72.14.232.152
12 53 ms 52 ms 51 ms 216.239.41.161
13 42 ms 48 ms 41 ms ord37s07-in-f14.1e100.net [172.217.1.46]

Trace complete.

tracert test to game's data center ip (during issue time frame)
Tracing route to 204.2.229.100 over a maximum of 30 hops

1 1 ms 1 ms 1 ms 192.168.1.1
2 9 ms 8 ms 9 ms 96.120.26.221
3 9 ms 8 ms 9 ms te-0-7-0-19-sur01.mendota.il.chicago.comcast.net
[162.151.34.109]
4 46 ms 48 ms 53 ms te-18-10-cdn04.mendota.il.chicago.comcast.net [6
9.139.202.154]
5 60 ms 47 ms 46 ms te-0-1-0-4-0-ar01.elmhurst.il.chicago.comcast.ne
t [68.86.189.41]
6 47 ms 55 ms 46 ms be-33491-cr01.chicago.il.ibone.comcast.net [68.8
6.92.33]
7 53 ms 51 ms 50 ms be-10506-cr02.350ecermak.il.ibone.comcast.net [6
8.86.86.229]
8 52 ms 50 ms 46 ms be-10577-pe03.350ecermak.il.ibone.comcast.net [6
8.86.86.2]
9 50 ms 46 ms 49 ms ae-26.a02.chcgil09.us.bb.gin.ntt.net [129.250.66
.65]
10 92 ms 91 ms 100 ms ae-4.r08.chcgil09.us.bb.gin.ntt.net [129.250.4.1
57]
11 42 ms 34 ms 40 ms ae-0.r21.chcgil09.us.bb.gin.ntt.net [129.250.2.2
05]
12 94 ms 87 ms 103 ms ae-5.r22.snjsca04.us.bb.gin.ntt.net [129.250.5.1
7]
13 104 ms 100 ms 95 ms ae-19.r01.snjsca04.us.bb.gin.ntt.net [129.250.3.
27]
14 91 ms 89 ms * ae-1.a00.snjsca04.us.bb.gin.ntt.net [129.250.2.2
28]
15 85 ms 91 ms 87 ms xe-0-0-42-1.a00.snjsca04.us.ce.gin.ntt.net [129.
250.195.42]
16 100 ms 110 ms 103 ms 204.2.229.242
17 102 ms 99 ms 101 ms 204.2.229.100

Trace complete.

pictures below (modem signals, nothing in logs)
qPGZvcg.png


router and modem
yWMoy3i.jpg


up close modem
4Fun1eY.jpg


wall coax cable panel
G2yV913.jpg


zoom out of coax cable panel area
eh9rYQ9.jpg


2 way splitter (this coax cable I figured out is for the tv in living room, we don't use it since we have no tv service at all).
QIwW56V.jpg


main 4 way splitter 3/4 bedroom wall panels come from here.
Zlttutd.jpg


basement ethernet cable (not sure if too close to other electrical wires?)
PaMTQ25.jpg


zoomed out view of that cable.
NZ0knrX.jpg


I'll take pictures of the outside where the main internet cable comes from the power line to the house, along with how the side of the house connections look, tomorrow while the sun is out.
 

vensus

Distinguished
Apr 28, 2016
86
0
18,640


Think I forgot to mention in my first post, but the 2 way splitter that came from the main splitter was removed already. I figured out that that whole coax cable leading into the 2 way splitter lead into our living room for that tv. I'm wondering if I call a tech to come out if they'd give me any issues if I tried explaining to them and asking if they could perhaps just have the main drop line that leads from the street go straight to the side of the house and into the room that hosts our modem/router, by passing any splitters and F to F adapters. It would make sense to do that in our case due to not using any TV service and we really only need one main cable for our services with internet.

I can remember having issues with explaining to comcast techs at our old place that once we got rid of their tv services, we didn't have any use for any of the other coax cables there either. (we had a main drop cable lead into a splitter that split into 3-4 other cables). No matter how much I told them that we only need the main drop cable, no splitters or extra cables at all because we DON'T use tv service at all! They would mess around with stuff outside and after checking the cables there would still be a splitter.. one tech finally told me that the splitter was there only to drop the signal a bit since it was running a bit too hot without one (think it was a -6db at that place).

I think our signal levels are a bit more balanced here from what I've seen so far, but with all the splitters and extra cables it's hard for me to think if I should ask for a direct drop line without a splitter to drop the signal slightly or if they'll just add a splitter regardless while they're checking the levels anyways. While a 4 way splitter is a bit of a waste just for a signal drop with one line it did work out at the old place, after the mess of having to remove anything else on that splitter off of it myself..

(slight rant, sorry) How should I proceed with a tech to come out and tell them to run the direct line drop to the room by passing all adapters and splitters? Maybe if the signal is too strong they can attach their 4 way splitter they love using to drop signal strength instead of a -6db piece, etc. Then from there watch an eye on things and see if anything changes for better or worse? (thank you for the help btw)

edit: I just checked ping levels on a few different things and seems everything is back to their avg ping again so as of right now it's looking like from the time frame of 7pm-11:45pm? cst. I'll have to monitor it tomorrow when I get up till a bit before 7pm and see if there's a pattern.
 

smashjohn

Reputable
Aug 14, 2017
574
12
5,365
In my old house they had to add a single splitter for the same issue (to lower the db). In my new house, they actually came over just to disconnect all our unused cables because it was jacking up the signal to the neighbors.

I wouldn't worry about hop 16 on your tracert. That's beyond Comcast. You might try a line quality test from DSL Reports to see if anything is flaky on Comcast's side. I think it runs for 24 hours.

I've heard you can set your starting frequency higher in order to push up into channels that fewer neighbors are using. In theory this could result in a better connection. But i've never messed with it.

Here's what my Router looks like. I'm pretty sure we have the same one. I'm not cable expert, but your numbers look good to me.

Procedure Status Comment
Acquire Downstream Channel 567000000 Hz Locked
Connectivity State OK Operational
Boot State OK Operational
Security Enabled BPI+
IP Provisioning Mode Honor MDD honorMdd(4)

Downstream Bonded Channels
Channel Lock Status Modulation Channel ID Frequency Power SNR Correctables Uncorrectables
1 Locked QAM256 13 567000000 Hz 3.2 dBmV 38.6 dB 3141 1937
2 Locked QAM256 1 489000000 Hz 3.8 dBmV 39.2 dB 2781 2010
3 Locked QAM256 2 495000000 Hz 4.1 dBmV 39.1 dB 767 1544
4 Locked QAM256 3 507000000 Hz 3.6 dBmV 38.3 dB 558 937
5 Locked QAM256 4 513000000 Hz 3.3 dBmV 38.8 dB 581 1021
6 Locked QAM256 5 519000000 Hz 2.5 dBmV 38.5 dB 606 1209
7 Locked QAM256 6 525000000 Hz 3.2 dBmV 38.7 dB 582 1444
8 Locked QAM256 7 531000000 Hz 2.3 dBmV 38.3 dB 1516 2106
9 Locked QAM256 8 537000000 Hz 2.4 dBmV 38.3 dB 1587 2112
10 Locked QAM256 9 543000000 Hz 2.7 dBmV 38.4 dB 467 1159
11 Locked QAM256 10 549000000 Hz 2.2 dBmV 38.3 dB 1763 1951
12 Locked QAM256 11 555000000 Hz 3 dBmV 38.5 dB 3330 2278
13 Locked QAM256 12 561000000 Hz 3.1 dBmV 38.3 dB 550 1250
14 Locked QAM256 14 573000000 Hz 4 dBmV 38.9 dB 2797 1877
15 Locked QAM256 15 579000000 Hz 3.8 dBmV 38.8 dB 737 1028
16 Locked QAM256 16 585000000 Hz 3.8 dBmV 38.7 dB 2729 1983
17 Locked QAM256 17 591000000 Hz 4.2 dBmV 39.3 dB 0 0
18 Locked QAM256 18 597000000 Hz 3.9 dBmV 39 dB 0 0
19 Locked QAM256 19 603000000 Hz 4.2 dBmV 39.1 dB 0 0
20 Locked QAM256 20 609000000 Hz 4.5 dBmV 39.4 dB 0 0
21 Locked QAM256 21 615000000 Hz 4.3 dBmV 39.1 dB 0 0
22 Locked QAM256 22 621000000 Hz 4.2 dBmV 39.2 dB 0 0
23 Locked QAM256 23 627000000 Hz 4.3 dBmV 39 dB 0 0
24 Locked QAM256 24 633000000 Hz 4.5 dBmV 39 dB 0 0

Upstream Bonded Channels
Channel Lock Status US Channel Type Channel ID Symbol Rate Frequency Power
1 Locked ATDMA 1 5120 Ksym/sec 37800000 Hz 38.5 dBmV
2 Locked ATDMA 2 5120 Ksym/sec 31400000 Hz 39.5 dBmV
3 Locked ATDMA 3 5120 Ksym/sec 25000000 Hz 39 dBmV
4 Locked ATDMA 4 5120 Ksym/sec 18600000 Hz 39 dBmV
5 Not Locked Unknown 0 0 Ksym/sec 0 Hz 0.0 dBmV
6 Not Locked Unknown 0 0 Ksym/sec 0 Hz 0.0 dBmV
7 Not Locked Unknown 0 0 Ksym/sec 0 Hz 0.0 dBmV
8 Not Locked Unknown 0 0 Ksym/sec 0 Hz 0.0 dBmV
 

vensus

Distinguished
Apr 28, 2016
86
0
18,640


Yeah I don't know what it is with them wanting to use splitters to drop signal instead of something else (think there's a piece that does just that without using extra wasted output/inputs). It would look like my router is about around the same levels so that's helpful to know nothing looks out of place there hopefully. I think I'll make a call out to comcast and hope I get a good tech to come out with them knowing I wish to remove a bunch of extra splitters/cables/etc during their time out here. Along with maybe seeing if that drop cable from the street is long enough to just travel a bit upwards and directly into the wall either by passing the F to F adapter or if they'd suggest leaving that in place so it isn't a direct cable from the power line to modem.

I'll still take a few pictures tomorrow with sun light of the outside cable setup and see what the soonest I can get a tech to come out.

edit: Also with it being about an hour now, I can safely say it seems that my ping must be only affected right around 7pm up till just 10-15mins before midnight cst. All pings and speeds seem to be stable as they normally should even with everyone in the house still doing what they were on devices a few hours ago. I assume it could also be the traffic of internet use of the neighborhood that happens in that time frame? But I hope there's something else to it then just that since it'd kinda suck to have gaming affected during that time frame from now on (maybe if anyone has been through that and knows how to fix it, etc can chime in?)
 

smashjohn

Reputable
Aug 14, 2017
574
12
5,365
I said line quality in my last post.. I meant run a Smoke Ping. This should monitor the latency of each hop over time. That hop #4 looks like a problem (te-17-10-cdn04.mendota.il.chicago.comcast.net [69.139.202.150]). I've had this issue with Comcast before, and getting their techs to take your reports seriously is really difficult. You'll probably have more luck with the onsite tech than with the phone tech. But to me it looks like a problem in comcast's network and not on your line.
 

vensus

Distinguished
Apr 28, 2016
86
0
18,640
Gotcha, I'll try running a smoke ping test and let it run for a day to see how things look. I've got another ping plotter test running now to google.com but part of me is concerned about why hop 2 and 3 area is getting a few ping spikes (I'll upload that test here within an hour or so when I get back to the house).

Took a few pictures of the drop line and where it leads, I'll upload those in a bit as well.. not an expert with coax setups but it looks like some lazy work? Few cables in splitters that they just cut short and left them there, etc.
 
Cable internet is "shared" medium - you're sharing it with all of your neighbors connected to the same cable (until that cable reaches ISP' communication box somewhere down the street). A couple of teenagers next door starting their torrents over the night are more than enough to break your pings.
 

vensus

Distinguished
Apr 28, 2016
86
0
18,640


Gotcha.. not really use to that if that's really the case here, in our other houses before this one. we lived outside of city limits and maybe shared internet with a neighbor or two, three tops but I never noticed anything weird like this. Would this be something you just leave alone and accept what happens during these time frames each day or is there anything I could do even with asking comcast to maybe upgrade our node to allow for bandwidth or something?

on a side note my pingplotter test to google.com ran for about over 2-3 hours and I'll post the results of that now.. Is the 2nd hop a major concern with the max pings spiking pretty high? I assume the 2nd hop is the connection leading through the neighborhood to the isp's router?

https://imgur.com/a/1Q9cm

pingplotter google test
https://i.imgur.com/MqL1Wgb.png
 
I would not be concerned about the physical connections. Your numbers are very good...in particular you are seeing very few errors.

Problems with the cabling will not show as higher latency it show packet loss. If the data is damaged it is discarded. Delays are caused by something holding the traffic in a memory buffer. This is almost always caused by too much data to process in a short time so you get a queue. There is also delays purely because of distance but that delay will not change from time to time.

You can easily share the cable with 100 houses around you so it is not just your direct neighbors. There is little you can do if the delay is outside your house and it is not you running too much traffic yourself.
 

vensus

Distinguished
Apr 28, 2016
86
0
18,640


Alrighty, yeah.. sadly with my history of comcast it's a major pain to get anyone higher in the ladder that will help fix anything (at least around our area they're not known for being that great with tech support). I'll give it a shot I guess and see if I can get in contact with the higher ups after back and forth techs if I must.



I would not be concerned about the physical connections. Your numbers are very good...in particular you are seeing very few errors.

{{Problems with the cabling will not show as higher latency it show packet loss. If the data is damaged it is discarded. Delays are caused by something holding the traffic in a memory buffer. This is almost always caused by too much data to process in a short time so you get a queue. There is also delays purely because of distance but that delay will not change from time to time.

You can easily share the cable with 100 houses around you so it is not just your direct neighbors. There is little you can do if the delay is outside your house and it is not you running too much traffic yourself.}}

Gotcha, thanks for the in depth info about how that works exactly. So what I'm piecing together is that everything within my house for setup should be fine and if I wish to proceed with trying to get any help for whats happening, that it'll be fully on comcast to do what they need either on the line right outside of my house on the street or even a few blocks away, etc. Would it still be ideal that if we're not using any other services except internet in our household, to see if we can just run the main drop line directly to our modem and if they must use anything extra at most like one splitter for drop signal a bit or one F to F adapter?

Otherwise it's just a lot of unused splitters/coax cables/etc added onto the line that won't be used.

(thank you all for the help again, I'm very greatful for any help!)


 
That seems to show a different issue than you described. In this case you show very consistent ping times but you are show a significant amount of packet loss.

So I am not sure what to say it may be a issue with the tool. Packet loss you would have to try to find the node that is dropping it. You can either manually leave ping run in a bunch of cmd windows to different nodes in the trace or you could try pathping. Pathping is simpler in some ways but it only runs for a short time so it the problem is not happening at that moment you will not see it.

If the loss was between your modem and the ISP you would see lots of uncorrectable errors. It means the problem farther into the network.

This testing method would also show you the hop that increases the latency even if you do not get loss.

Still time of day issues are almost always related to load. Some rare issues are caused by heating and cooling of equipment but the vast majority of the time it is all the people coming home from work and turning on netflix or whatever.

It is kinda like steam game downloads. If I do it at 3am it will get 350mbps and my ISP only promises 300. If I try it at 7pm especially on the weekend I might get 50mbps. This can be either steam or the ISP but it is definitely my traffic competing for bandwidth.
 

vensus

Distinguished
Apr 28, 2016
86
0
18,640


I don't think I'm getting any packet loss that's too major, what loss that does show I think seems to come from the final hops when I've ran pingplotter to google and some game servers, etc. I also noticed that yesterday during that timeframe I was having a 20-30+ ping increase didn't happen at all as I kept checking it up till I went to sleep around 1-2am. I think my post on dslreports, someone suggested running that smokeping test just to check things out (sadly I can't really read it or understand it much).

Think my first course of action will be getting a tech out here soon as possible to run the direct line to my modem and free up any extra stuff attached to everything sometime next week, then monitor and run a few more tests with it setup like that and see whats the next course of action.
 

vensus

Distinguished
Apr 28, 2016
86
0
18,640
Last update for a while, called comcast today to get a tech to come out and the lady on the phone didn't seem to quite understand what I meant by having the direct drop line go straight to the modem, by passing any splitters and other cables. The best thing they could say that I figure would do the job right in another way is that they suggested installing a new line to the house and disabling the old line instead but at the cost of $60.

I'm still going to chat with the tech who comes out before he starts and see if he can understand what I'm explaining better in person before installing another line and paying out for it. But even if I do get stuck with that option I assume that'd be good still long as the new line is from the street line where the other one comes from and have the new one go directly to modem still? I won't be able to update much since the tech won't be out here till the 21st of jan.. but I'll update on how things went and how things are looking after the visit.