jfm at blueyonder.co.uk
Wed Jun 6 02:35:50 BST 2001
Paul Civati <paul at xciv.org> wrote:
>John Murphy <jfm at blueyonder.co.uk> wrote:
>> >What test did you perform to determine this 30-40% loss?
>> FreeBSD ping to the 10.48.0.1 router.
>% ping -c 100 -q 10.87.144.1
>PING 10.87.144.1 (10.87.144.1): 56 data bytes
>--- 10.87.144.1 ping statistics ---
>100 packets transmitted, 100 packets received, 0% packet loss
>round-trip min/avg/max/stddev =3D 10.311/11.568/13.905/0.769 ms
>It's possible that it could be rate limiting ICMP responses.
It's been fine all evening, until I sent them a nice thankyou e-mail.
Now it's not as bad as it was (about 12% packet loss).
>Whilst doing the above ping, I did find that it timed out on
>that hop whilst doing a traceroute, so perhaps some kind of rate
>limiting going on there, but I don't think anything that would
>affect normal traffic.
Possibly because there's no dns for the first hop. I use ping and
traceroute with -n for these tests.
>I would test by pinging a host inside and outside the BY network,
>will help you track down if/where there is a problem.
I left ping running every 5 minutes during peak time thisevening
(-c 16) and got 0% packet loss 'till 1am. approx. So the load
doesn't seem to affect the results.
I don't think I'll be going back to dial-up though. It's great
when it's working...
More information about the Ukfreebsd