Skip to content
elih's profile

New Member

 • 

3 Messages

Wednesday, May 21st, 2014 10:00 PM

Bad routes

We support several small businesses in the area and have noticed bad routes for some of our clients, this one is particularly weird, see hop 6.  Is there anything that can be done about this?  I can PM source and destination IPs if needed.

 

1.|-- **.***.***.** 0.0% 10 0.1 0.1 0.0 0.1 0.0
2.|-- 10.65.7.5 0.0% 10 0.3 0.4 0.3 0.7 0.0
3.|-- te-8-4.car2.Denver1.Level 0.0% 10 0.8 73.8 0.8 203.7 78.9
4.|-- ae-1-51.edge3.Denver1.Lev 0.0% 10 0.7 0.7 0.7 0.7 0.0
5.|-- comcast-level3.Denver1.le 0.0% 10 3.0 3.0 1.3 4.6 1.1
6.|-- pos-0-14-0-0-cr01.miami.f 90.0% 10 2.5 2.5 2.5 2.5 0.0
7.|-- he-4-11-0-0-cr01.sanjose. 0.0% 10 27.4 28.9 27.4 30.9 1.2
8.|-- he-0-6-0-0-ar01.sfsutro.c 0.0% 10 31.2 30.6 28.8 32.3 0.9
9.|-- te-0-4-0-6-sur03.sanjose. 0.0% 10 30.7 30.9 30.6 31.7 0.0
10.|-- te-6-0-acr13.sanjose.ca.s 0.0% 10 31.7 31.7 31.7 31.8 0.0
11.|-- c-76-103-248-144.hsd1.ca. 0.0% 10 38.3 39.4 38.0 45.6 2.2
12.|-- **-***-***-***-static.hfc 0.0% 10 38.9 39.3 38.6 41.6 0.8

 

 

 

 

 

 

Accepted Solution

Visitor

 • 

3 Messages

11 years ago

That might not actually in in Miami.  The latency doesn't indicate the packet having to actually go that far.  It's probably just an incorrect PTR record for the router's interface IP address.

Accepted Solution

Problem solver

 • 

117 Messages

11 years ago

ELIH,

 

Traceroute results are always interesting.  Traceroute uses tailored pings with increasing hop count for each query. Intermidiate responses come from routers along the path where the hop count expires. The last response comes from the target system. The significant conclusions from the results you posted include:

  

1. The miami response may indicate that the router queried was too busy forwarding data packets to handle ICMP Echo requests.  Also, the actual device location may or may not be in some miami somewhere. Network names are not always created for customer convenience.

 

2. The 39 ms Round Trip Time  to hop 12 is quite reasonable.  If this is your target system, then you have no real problem with this path.

 

Last, since IP routing is dynamic, the results of a traceroute does not indicate a static route, just a snapshot.  The prime conclusion from your posted is that the path tested us satisfactory for all ordinary traffic.

Problem solver

 • 

305 Messages

11 years ago

Generally loss in the middle of an MTR is nothing to be concerned about. You'll want to look at the end of the MTR (which looks clean for you). I'd also recommend running a reverse MTR if at all possible.

New Member

 • 

3 Messages

11 years ago

I was looking more at the Denver -> Miami -> San Jose route, seems a bit of a long way to get from CO to CA through FL.  Reverse MTR looks fine.

 

Thanks for the quick reply though.

New Member

 • 

3 Messages

11 years ago

Thanks for the replys guys, always been blindly trusting those names, you're right the latency is about what I'd expect.