dslreports logo
site
 
    All Forums Hot Topics Gallery
spc

spacer




how-to block ads


Search Topic:
uniqs
2532
share rss forum feed


NetEng

@windstream.net

[Network] Core network latency issue 98.16.195.26 (Bad EGIRP rou

Afternoon Windstream, seems once again a few hops upstream of the BRAS you've got significant congestion affecting both Business and Residential DSL circuits and EGIRP paths. Issue began around 11:10 CDT.

98.16.195.26 is seeing significant latency. This is nearly an exact repeat of »Transport/Network - 3rd hop past BRAS/AC extreme latency

Linux traceroute, as below, uses UDP by default, not ICMP.

Business DSL Circuit
# traceroute www.google.com -nn
traceroute to www.google.com (173.194.74.103), 30 hops max, 38 byte packets
 1  173.186.112.146  28.169 ms  28.606 ms  27.396 ms
 2  173.187.47.49  29.077 ms  29.168 ms  28.921 ms
 3  98.16.195.26  205.012 ms  199.791 ms  197.944 ms
 4  166.102.72.88  260.137 ms  162.39.222.102  193.580 ms  272.099 ms
 5  166.102.72.13  186.305 ms  72.14.219.192  193.592 ms  166.102.72.13  199.273 ms
 6  64.233.174.2  203.906 ms  72.14.239.100  198.037 ms  64.233.174.2  194.091 ms
 7  66.249.94.6  195.960 ms  66.249.94.20  200.689 ms  66.249.94.6  192.812 ms
 8  72.14.239.65  208.107 ms  216.239.48.5  213.829 ms  216.239.48.41  216.791 ms
 9  72.14.234.55  217.804 ms  72.14.234.53  214.635 ms  213.797 ms
10  *  *  *
11  173.194.74.103  219.874 ms  227.295 ms  212.148 ms
 

Residental DSL Circuit
# traceroute www.google.com -nn
traceroute to www.google.com (74.125.131.103), 30 hops max, 38 byte packets
 1  173.191.128.1  40.068 ms  34.504 ms  34.755 ms
 2  173.187.47.61  34.332 ms  34.904 ms  34.917 ms
 3  98.16.195.26  186.359 ms  193.737 ms  186.748 ms
 4  166.102.72.88  182.977 ms  162.39.222.102  180.331 ms  166.102.72.88  215.263 ms
 5  72.14.219.192  170.190 ms  183.828 ms  71.28.137.173  187.483 ms
 6  72.14.239.100  176.783 ms  171.492 ms  200.554 ms
 7  66.249.94.22  182.208 ms  66.249.94.20  183.708 ms  178.047 ms
 8  209.85.249.238  182.239 ms  186.949 ms  202.293 ms
 9  216.239.48.103  167.726 ms  64.233.174.117  164.129 ms  167.728 ms
10  *  *
 


NetEng

@windstream.net

Re: [Network] Core network latency issue 98.16.195.26 (Bad EGIRP

Resolved -- I also was able to put a ticket in and request transport/net eng see it and it was related to the "Latency outage issue on your DSLAM" (which isn't accurate).

Apologies for the typo on EIGRP, as you can see 98.16.195.26 is no longer in my route.

Residential DSL
# traceroute -nn www.google.com
traceroute to www.google.com (74.125.131.106), 30 hops max, 38 byte packets
 1  173.191.128.1  33.488 ms  33.792 ms  33.907 ms
 2  173.187.47.57  36.937 ms  35.017 ms  34.726 ms
 3  173.187.47.233  46.770 ms  45.188 ms  43.587 ms
 4  166.102.72.80  42.858 ms  45.455 ms  72.193 ms
 5  166.102.72.13  42.628 ms  43.184 ms  83.100 ms
 6  64.233.174.2  46.564 ms  45.411 ms  50.017 ms
 7  66.249.94.24  47.262 ms  66.249.94.6  49.707 ms  49.291 ms
 8  209.85.249.45  57.283 ms  209.85.249.238  57.755 ms  60.835 ms
 9  216.239.48.103  57.901 ms  209.85.251.228  61.238 ms  60.963 ms
 


NetEng

@windstream.net
said by NetEng :

I also was able to put a ticket in and request transport/net eng see it and it was related to the "Latency outage issue on your DSLAM" (which isn't accurate).

This sentence fails to communicate what I'm trying to say. There is often times an existing wildcard ticket applied to all accounts in my area that is a "Latency outage on your DSLAM" which is not accurate, the DSLAM is able to service the subscribers without issue. Often, this ticket has prevented me from routing accurate tickets/outage reports to the correct places, or even logging a ticket at all.

In today's instance, the CSR was able to correctly apply a ticket and notate the true outage, far upstream in the Windstream network, perhaps due to an errant EIGRP push or routing table. I did not mean to insinuate that today's outage/issue was blamed on the "Latency outage on your DSLAM" super-ticket.

nOv1c3

join:2006-11-08
Whitney, TX
kudos:1
reply to NetEng
I feel your Pain , I,m in the same boat , Its their whole network that needs upgrading ,

My latency in none peak hours is 20ms to first hop after about 5 or so my ping jumps to 60+ to first hop second hop kicks it up 180+ms third and fourth gets worse ,, All on windstream nodes

Along with the ping spikes we get the added bonus of packet lost just to top it off lol


NetEng

@windstream.net
Respectfully, I disagree. This was an isolated incident that was corrected in 4 hours. I have no issues aside from these isolated incidents and consistently get 100% of the bandwidth I'm provisioned for, on both circuits.

Additionally, I couldn't care less what "Speedtest" things my "Grade" is, it's a ridiculous metric for measuring network performance.

Upstream QoS set at 600K


nOv1c3

join:2006-11-08
Whitney, TX
kudos:1
reply to NetEng
I,m Glad its just a isolated incident , But as you can see all over these forums this has been an ongoing issue for years

Keep an eye on it , Once it starts degrading it well just get worse and worse , we have all seen it

Atekido

join:2002-08-25
Crossett, AR
reply to NetEng
There is no such thing as Isolated with WS...