 | reply to frivolous
Re: Routing Problems Perfect. Thank you!
First of all, the packet-loss (shown in red) is patterned which indicates some form of ICMP rate limiting, likely on your router. From the CMTS (hop 2) to the private IP of the cable modem and it shows 0 packet-loss a s seen below:
CMTK01#ping {removed} repeat 5000
Type escape sequence to abort. Sending 1000000, 100-byte ICMP Echos to {removed}, timeout is 2 seconds: Success rate is 100 percent (5000/5000), round-trip min/avg/max = 4/8/10 ms
Second, according to the PingPlotter graph, the latency is occurring between hop 2 and 3 and only occurs between 6pm and 11pm. Let me dig deeper into that and figure out what's going on with that. I'll PM you on here when I find out. |
|
 | thank you for your help, is there anything i can do to help the packet loss? or no? - thanks again! |
|
 | reply to frivolous Since it's just rate-limiting ICMP, then it's really just cosmetic and should not affect any real traffic.
Edit: You can either figure out where the setting on your router is to disable the rate-limiting or you can temporarily bypass your router and run pingplotter while directly connecting your computer to your cable modem. This should remove the rate limiting from the PP trace.
There could be some real packet-loss in there but since the rate limiting is already causing so much packet-loss, it may be hiding it. |
|
 | Change PP to use different protocol? Noticed the setting early, never tried it to see what happens, lol |
|
 | reply to frivolous I have isolated the latency problem and we will be correcting it on Thursday morning barring any unforeseen delays. |
|
 | Out of curiosity, why are you able to single handedly do more for everyone on this forum than all the call centers combined? |
|
 gatorkramNeed for SpeedPremium join:2002-07-22 Winterville, NC kudos:2 | said by moldypickle:Out of curiosity, why are you able to single handedly do more for everyone on this forum than all the call centers combined? My guess, is the fact in his name, its level 3, when you call the call center, thats like level 0 to 1, and they send stuff to level 2 if you are lucky, and then the level2s end up sending stuff to level 3 again, if you get lucky.
This is my total guess. I also don't think there is a level 4  -- What the heck is a GatorKram? »www.gatorkram.com |
|
 | reply to frivolous Actually, the L3 stands for "layer 3" which is the network layer of the OSI model (»en.wikipedia.org/wiki/OSI_model).
I do not work for a call center. I am a routing engineer. |
|
 gatorkramNeed for SpeedPremium join:2002-07-22 Winterville, NC kudos:2 | D'oh! Could I have been more wrong...  -- What the heck is a GatorKram? »www.gatorkram.com |
|
|
|
 | reply to SDL L3Tech Just want to say that SDL L3Tech has been such a huge help to many in this forum... he/she deserves a nice raise in salary for going above and beyond. Should we start a petition? |
|
 | reply to Moostang Looking forward to the fix tomorrow i nthe am, Moostang suggested running a trace route back to my ip - here it is
1 0.1 ms 0.1 ms 0.1 ms router.internap-la.nfoservers.com [64.94.101.254] 2 0.6 ms 0.3 ms 0.3 ms border2.te9-4.nuclearfallout-53.lax010.pnap.net [63.251.209.181] 3 0.5 ms 0.5 ms 0.4 ms core1.po2-20g-bbnet2.lax.pnap.net [216.52.255.65] 4 0.8 ms 0.8 ms 0.7 ms xe-0-3-0-1.r04.lsanca03.us.bb.gin.ntt.net [198.172.90.25] 5 0.5 ms 0.5 ms 0.5 ms xe-1.level3.lsanca03.us.bb.gin.ntt.net [129.250.9.202] 6 8.7 ms 12.9 ms 8.6 ms vlan80.csw3.LosAngeles1.Level3.net [4.69.144.190] 7 20.8 ms 9.0 ms 8.9 ms ae-83-83.ebr3.LosAngeles1.Level3.net [4.69.137.41] 8 8.9 ms 8.9 ms 8.9 ms ae-3-3.ebr1.SanJose1.Level3.net [4.69.132.9] 9 13.4 ms 8.8 ms 16.2 ms ae-71-71.csw2.SanJose1.Level3.net [4.69.153.6] 10 8.8 ms 8.9 ms 8.9 ms ae-2-70.edge2.SanJose1.Level3.net [4.69.152.79] 11 9.4 ms 9.4 ms 9.3 ms SUDDENLINK.edge2.SanJose1.Level3.net [4.79.238.38] 12 19.3 ms 19.4 ms 41.3 ms 173-219-251-155-link.sta.suddenlink.net [173.219.251.155] 13 17.8 ms 17.8 ms 17.8 ms 173-219-251-225-link.sta.suddenlink.net [173.219.251.225] 14 * * * Request timed out. 15 * * * Request timed out. 16 * * * Request timed out. 17 * * * Request timed out. 18 * * * Request timed out. 19 * * * Request timed out. 20 * * * Request timed out. 21 * * * Request timed out. 22 * * * Request timed out. 23 * * * Request timed out. 24 * * * Request timed out. 25 * * * Request timed out. 26 * * * Request timed out. 27 * * * Request timed out. 28 * * * Request timed out. 29 * * * Request timed out. 30 * * * Request timed out.
Trace complete.
**On a side note - A raise is in order for all the work you have done! |
|
 | But still, how is it you are able to fix probably 80% or better of the problems of this whole forum, yet we haven't had literally 1 person on here saying ANY previous attempt at contact w/ CS has worked.....
It's even more mind blowing when you inform us you're not even really in the CS chain,,,, but still fix the shit for us (or get it fixed).
Probably wouldn't be so bad if there were even 1 in 12 phone support people that had a clue wtf they were talking about :-- 30/2 Suddenlink : Current 5/1 CMA : Old 15/2 TWC : Old |
|
 | reply to frivolous Prolly has more to do with access. I'm betting they don't let their call center agents have access to the core network.
I mean i'm sure they have their own gui based diagnostic tools but they're likely based on checking the modem if the agents even know what the data means. But I doubt it shows the utilization of the network between the CMTS and the internet and if it did, what could an agent really do other than escalate it.
In my experience, call center agents aren't trained in in-depth troubleshooting techniques on a nationwide network but rather are highly trained in processing the sheer customer driven call volumes. |
|
 | Kinda half my point there in what you said. The call center people aren't trained, don't know what the data even means that they DO have in front of them, and I have never once been escalated (just been told that they would). When I call in anymore I have utterly given up on asking for help and just simply ask them to check if there are any reported issues or maint. going on in my area.
Access and data aside, why is it that the call center people can't even get basic help or at the very least the ball rolling into another department/'help tier'. L3 tech here seems to be able to do that much (Pete certainly is able to, LOL)
Off topic I guess, so /rant. -- 30/2 Suddenlink : Current 5/1 CMA : Old 15/2 TWC : Old |
|
 | reply to frivolous frivolous - The maintenance was completed and is looking good so far. Run PP for the next 24 and see if it's better. |
|
 | Again, thank you very much, i will run pp and load up the results tomorrow. |
|
 jdmm72 join:2002-02-12 Nitro, WV Reviews:
·Suddenlink
| reply to frivolous Most call centers are measured on first call resolution (FCR). It is in their interest to close the calls out, as opposed to escalating to the next level. I used to work at an internal helpdesk for US HUD in 2006. We weren't really measured on this metric, and had much better access to tools and other groups (but we were also internal), but most of the people working at the "tech support" aren't required to even have a background in IT at all, just read a script. We were required to have MCDST, and MCP certifications within 6 months. Yes, getting MCDST also got you MCP. We were encouraged to have A+, Network+, and CCNA, and any other we wanted. One guy actually decided to go through ethical hacking, and was approved. The company (EDS through CNSI, now HP) would pay for certifications.
In 2000-2001, I worked for BellSouth ADSL for tech support. I actually was working on and eventually had my CCNA cert (not common in 2001), but would get in trouble (well, not trouble, just not good metrics) a lot for deviating from script and actually getting people fixed. I called the NOC a lot, had long calls, but got people fixed. Short calls and first call resolution were the most important metrics, not actual fixes. AT&T paid our company based on number of calls, and FCRs. |
|
 | reply to SDL L3Tech
Here is that new PP : |
|
 | reply to frivolous The latency looks clean now. Notice it showed no change during peak-time like your previous PP trace. |
|
 | reply to frivolous Awesome - thank you very much for your time! |
|