Re: Mauritius Internet Exchange Point Adventures

From: Loganaden Velvindron <loganaden_at_gmail.com>
Date: Fri, 23 Oct 2015 19:01:34 +0000

On Fri, Oct 23, 2015 at 6:52 PM, S Moonesamy <sm+mu_at_elandsys.com> wrote:

> Hi Logan,
> At 09:40 23-10-2015, Loganaden Velvindron wrote:
>
>> Here's a follow-up with some technical part:
>>
>
> The ping output is not displaying correctly.
>

Fixed. Refresh your cache.


>
> Here's my output:
>
> PING 154.71.1.18 (154.71.1.18): 56 data bytes
> 64 bytes from 154.71.1.18: icmp_seq=0 ttl=247 time=5.285 ms
> 64 bytes from 154.71.1.18: icmp_seq=1 ttl=247 time=5.134 ms
> 64 bytes from 154.71.1.18: icmp_seq=2 ttl=247 time=5.066 ms
> 64 bytes from 154.71.1.18: icmp_seq=3 ttl=247 time=4.889 ms
> 64 bytes from 154.71.1.18: icmp_seq=4 ttl=247 time=4.934 ms
> --- 154.71.1.18 ping statistics ---
> 5 packets transmitted, 5 packets received, 0.0% packet loss
> round-trip min/avg/max/std-dev = 4.889/5.061/5.285/0.162 ms
>
> I don't have access to a FTTA connection to do the test in reverse [1]. I
> don't know why you are getting packet loss. I would not conclude that
> "more than 3/5 of the traffic is lost" because of the results of your
> test. You commented that

The high packet loss is worrying for wireless connections. It slows down
the performance, and degrades the Internet experience.

We need more tests to determine if it's only occasional, or occuring
systemmatically.




> there was "some action from Emtel and Orange to fix the peering". As a
> user I would agree that one or both of these companies did something
> positive.
>
>
I think that we need measurement tools that always display the latency back
and forth from different ISPs. That way, any user can notice it, and raise
the alart instead of the same people ringing the alarm each time.



> Regards,
> S. Moonesamy
>
> 1. The test is debatable.
>
Received on Fri Oct 23 2015 - 19:01:47 PST