Poor Connectivity, 10-60 second lag spikes and outages

The link for yesterday was wrong...

 http://vindicator.us/images/ComcastBusiness.10-23-2012-TWG.jpg

This is the correct link for 10-23-12

TheWebGamer.   While the outage issue was cleared last evening we are still monitoring the gateway for resets and errors.  If you will send me any of  the terminating IP's  that you are seeing failures to we can test thru the network path to locate the source of the issue.  

Thank you.

173.166.234.158 is directly on your SMC-D3 modem.  I test only to your modem, and never to any IP on my dise of the modem.

Called into Business Class Tech Support tonight at 11:30 PM PST.  All of the above CR and ER numbers were marked closed and resolved.  Wish someone would have called and told me that so I could laugh in their face.

The last 3 hours my average packet loss was 3.19%  Highest was 18.81 and it is currently 2.86%

The new CR number for tomorrow is CR302272701.  Tech #31333 offered me an ER number which I declined.  The last two days in a row I had a nice escalated ER number and that failed to even get anything done, so why bother.

I could post images of my modem signals etc, but they were as bad as last night... so same old news.  I did save them and document them in our files though.

Morning John!   What is the word from the inside about our issues?  

As for me, I am looking forward to another call from Britney and or Sanji from Comcast Corporate Escalations tomorrow to go over the past 3 days with them.  I have kept Britney in the loop with every ticket and the responce, or lack there of.

I will post more status reports tomorrow. Right now it is 00:30 AM Pacific, My internet dropped 20-30 ms of latency instantly and I have a nice solid green line with zero packet loss.  Just like someone just flips a switch and all the line noise goes away... every single night, at the same time.  You can set your watch on it.

Looking at the reports this morning.. 🙂 The top graph is what it should always look like.  Lets see how long it lasts.

William, We are still monitoring form the gateway upstream to the node. We are testing to the node as well as running a smokeping test from DSL reports.  The results are posted below.  

GATEWAY TO NODE  results

packet size 1024

--- 69.139.197.233 ping statistics ---
200 packets transmitted, 200 packets received, 0% packet loss
round-trip min/avg/max = 0/8/60 ms
Main>

--- 69.139.197.233 ping statistics ---

packet size 1400
500 packets transmitted, 500 packets received, 0% packet loss
round-trip min/avg/max = 0/8/80 ms
Main>

 I also checked smokeping result on DSLReports

CA1

As you are aware all results are dependent on traffic volume at the time of test.   

Comcast is continuing to investigate the issue of Latency from your gateway to the node.

Thank you

Ok Comcast John...  This one is for you.  Thank you for giving me the very very close to me IP address you ran those very very short tests from during at 10:07 AM my time at which time we hav ethe very best of what Comcast can offer.  You really need to read and look at the graphs to know we FAIL at night.  Waving that crap test under my nose like I have no idea what I am talking about insults me and makes you look foolish.

But what is more to the point is that you failed to set the correct date after you photo shopped the immage I suspect...  You posted on the 29 an image for October 31st, which was several days in the future...

Let me show you a real test using the same IP address information you so kindly pointed out.  Time of al lthe tests was 10:54 PM Pacific Time.  Date and time on DSLReports Smoke ping is Eastern Time and shows 00:45:49 10-31-2012

Lets start with my traffic that you are all trying to say is the problem...  4 whole users...  I have had 3000 users this summer with no lag.

Lets look at the ping route that you used to test my modem...  Looks like you are pinging me from Roseville Road at Elkhorn Blvd in Sacramento CA where the Head End is located?

Lets now look what DSLReports is saying from the smoke ping...

Now at the same time lets send 1000  256k packets with a 500ms timeout and see that tonight I only have 2% packet loss between here and the head end.   AGAIN!!!  I am complaining about my traffic lag AT NIGHT!!!  AT NIGHT!!!  AT NIGHT!!!

Pissing away your tech and our time in the morning out here after the problem goes away for 15 months does not solve this problem...   So your worthless test at 10:AM means nothing at all.  You just wasted your time just like all the other caring Comcast employees.

So there you go... We are having a good night here at only 2% packet loss...   Next time to a real test that counts when it counts.  And as for my bandwidth, you are monitoring it and know dam well I am not using very much at all.

WebGamer.  Thank you and good catch on the stats post.  The CA1 image was posted as an embedded live link which will allow for data updates by clicking the link, which also updates the time stamp. Ongoing we will post screen shots only. 

Thanks again for pointing this out.

Webmaster.  

Thanks again for all your time and attention to this topic.   Local operations and the Tier 2 support group have worked together to determine a solution for latency issues during high traffic times.  We show that you purchased Business Class Internet Deluxe with 50Mbps x 10 Mbps.  However, you consistently demand more than 50x10 from the circuit, and the consistent demands above the service levels are causing degradation of service.  If you consistently require more than 50x10, we recommend some of our higher bandwidth services which may be more appropriate to meet your demands.   From what we can tell, some of our enterprise services may be more appropriate.  If you want, we can connect you with an enterprise service rep to discuss what may be best to meet your bandwidth usage needs.

Thank You

Here is the direct effect the outage at the head end had upon our servers.  We did loose 2 more customers which have also been documented and clearly stated by our customer that connectivity was the issue again.

Looking at both images, you can clearly see that this issue had nothing to do with any connectivity or bandwidth issues between out company and our node at the Comcast head end...