How to analyze mtr data

Question

When I use to have latency issue with my hosting server I use to send mtr report to support so they can analyze and tell me if there is any issue at my end or their servers.

Now, I am running with issues where I need to transfer files to an on-line service. My speed test shows me i have very good internet speed 60Mbps but when I upload the files it upload at 32KBps. I first thought it could be an issue with their server (located in france) and they confirmed they have no such issues and their other clients have no problems. So it must be something something happening in between when I am transfering/uploading from India.

Below is my mtr report to www - ranchcomputing - com. I see there are lot of packet loss in the middle but How exactly should I read this and report to my manager and explain him the situation.

|------------------------------------------------------------------------------------------|
|                                      WinMTR statistics                                   |
|                       Host              -   %  | Sent | Recv | Best | Avrg | Wrst | Last |
|------------------------------------------------|------|------|------|------|------|------|
|                             192.168.1.1 -    0 |   32 |   32 |    0 |    0 |    2 |    1 |
|           1.128.204.49-ras.beamtele.net -    0 |   32 |   32 |    1 |    1 |    4 |    2 |
|                        ras.beamtele.net -    0 |   32 |   32 |    2 |   16 |  105 |    2 |
|                        ras.beamtele.net -    0 |   32 |   32 |    1 |    4 |   33 |    2 |
|121.240.252.1.STATIC-Hyderabad.vsnl.net.in -    4 |   32 |   31 |    1 |    3 |   13 |    5 |
|                   No response from host -  100 |   32 |    0 |    0 |    0 |    0 |   15 |
|   ix-0-100.tcore1.MLV-Mumbai.as6453.net -   32 |   32 |   22 |   14 |   19 |   92 |   14 |
|                           80.231.217.17 -   84 |   31 |    5 |  334 |  336 |  339 |  339 |
|   if-8-1600.tcore1.PYE-Paris.as6453.net -    4 |   31 |   30 |  332 |  341 |  365 |  346 |
|                   No response from host -  100 |   31 |    0 |    0 |    0 |    0 |   47 |
|________________________________________________|______|______|______|______|______|______|

Below is PathPing

Tracing route to ranchcomputing.com [213.186.33.4]
over a maximum of 30 hops:
  0  MYPC [192.168.1.107] 
  1  192.168.1.1 
  2  1.128.204.49-ras.beamtele.net [49.204.128.1] 
  3     *     ras.beamtele.net [183.82.15.5] 
  4     *     ras.beamtele.net [183.82.14.97] 
  5     *     121.240.252.1.STATIC-Hyderabad.vsnl.net.in [121.240.252.1] 
  6     *        *        *     
Computing statistics for 125 seconds...
            Source to Here   This Node/Link
Hop  RTT    Lost/Sent = Pct  Lost/Sent = Pct  Address
  0                                           MYPC [192.168.1.107] 
                                0/ 100 =  0%   |
  1    0ms     0/ 100 =  0%     0/ 100 =  0%  192.168.1.1 
                                0/ 100 =  0%   |
  2    1ms     0/ 100 =  0%     0/ 100 =  0%  1.128.204.49-ras.beamtele.net [49.204.128.1] 
                                0/ 100 =  0%   |
  3    4ms     0/ 100 =  0%     0/ 100 =  0%  ras.beamtele.net [183.82.15.5] 
                                0/ 100 =  0%   |
  4    2ms     0/ 100 =  0%     0/ 100 =  0%  ras.beamtele.net [183.82.14.97] 
                                0/ 100 =  0%   |
  5    2ms     0/ 100 =  0%     0/ 100 =  0%  121.240.252.1.STATIC-Hyderabad.vsnl.net.in [121.240.252.1] 

Trace complete.

Answer 1

How exactly should I read this and report to my manager and explain him the situation?

As you pointed out, you have issues towards the end of your mtr testing. Specifically these 2 hops are troubling for the amount of consistent packet loss during this test:

ix-0-100.tcore1.MLV-Mumbai.as6453.net

80.231.217.17

You should be able to make your manager aware of issues you have seen in troubleshooting through these tests. Of course you should run more extensive tests for longer periods of time. 32 count for attempts using mtr is relatively short.

Broader explanation of headers:

% means percentage of packets lost. (This is your primary issue it appears. More info about packet loss here.)

Sent is the count of attempts sent out.

Recv is the count of attempts responded or received back.

Best is the best millisecond count of latency you have seen during this testing.

Avrg is the average milliseconds of latency during this testing.

Wrst is the worst milliseconds of latency you have seen during this test.

Last is the most recent reading for milliseconds of latency during this test.