You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by Randy Terbush <ra...@zyzzyva.com> on 1997/06/26 21:43:42 UTC

Re: Hyperrouter's Reality

27% packet loss from here. The BBN-UUNET border router situation 
does not look good from here.

zab:/var/users/doc/www.sportdoc.com
#> ping -s 1500  twinlark.arctic.org
PING twinlark.arctic.org (204.62.130.91): 1500 data bytes
1508 bytes from 204.62.130.91: icmp_seq=0 ttl=55 time=150.896 ms
1508 bytes from 204.62.130.91: icmp_seq=1 ttl=55 time=150.455 ms
1508 bytes from 204.62.130.91: icmp_seq=2 ttl=55 time=146.558 ms
1508 bytes from 204.62.130.91: icmp_seq=3 ttl=55 time=136.520 ms
1508 bytes from 204.62.130.91: icmp_seq=5 ttl=55 time=139.806 ms
1508 bytes from 204.62.130.91: icmp_seq=8 ttl=55 time=148.004 ms
1508 bytes from 204.62.130.91: icmp_seq=9 ttl=55 time=133.583 ms
1508 bytes from 204.62.130.91: icmp_seq=10 ttl=55 time=137.575 ms
1508 bytes from 204.62.130.91: icmp_seq=11 ttl=55 time=135.580 ms
1508 bytes from 204.62.130.91: icmp_seq=12 ttl=55 time=146.005 ms
1508 bytes from 204.62.130.91: icmp_seq=14 ttl=55 time=146.152 ms
1508 bytes from 204.62.130.91: icmp_seq=16 ttl=55 time=144.001 ms
1508 bytes from 204.62.130.91: icmp_seq=17 ttl=55 time=133.629 ms
^?
--- twinlark.arctic.org ping statistics ---
18 packets transmitted, 13 packets received, 27% packet loss
round-trip min/avg/max = 133.583/142.212/150.896 ms


As opposed to my connection to taz:

#> ping -s 1500  apache.org
PING apache.org (204.152.144.38): 1500 data bytes
1508 bytes from 204.152.144.38: icmp_seq=0 ttl=249 time=105.282 ms
1508 bytes from 204.152.144.38: icmp_seq=1 ttl=249 time=106.956 ms
1508 bytes from 204.152.144.38: icmp_seq=2 ttl=249 time=105.661 ms
1508 bytes from 204.152.144.38: icmp_seq=3 ttl=249 time=111.128 ms
1508 bytes from 204.152.144.38: icmp_seq=4 ttl=249 time=108.928 ms
1508 bytes from 204.152.144.38: icmp_seq=5 ttl=249 time=108.576 ms
1508 bytes from 204.152.144.38: icmp_seq=6 ttl=249 time=103.666 ms
1508 bytes from 204.152.144.38: icmp_seq=7 ttl=249 time=105.187 ms
1508 bytes from 204.152.144.38: icmp_seq=8 ttl=249 time=121.232 ms
1508 bytes from 204.152.144.38: icmp_seq=9 ttl=249 time=120.758 ms
1508 bytes from 204.152.144.38: icmp_seq=10 ttl=249 time=103.432 ms
1508 bytes from 204.152.144.38: icmp_seq=11 ttl=249 time=112.013 ms
1508 bytes from 204.152.144.38: icmp_seq=12 ttl=249 time=110.520 ms
1508 bytes from 204.152.144.38: icmp_seq=13 ttl=249 time=103.341 ms
1508 bytes from 204.152.144.38: icmp_seq=14 ttl=249 time=107.021 ms
^?
--- apache.org ping statistics ---
15 packets transmitted, 15 packets received, 0% packet loss
round-trip min/avg/max = 103.341/108.913/121.232 ms


> Brian and I have arranged to move hyperreal to another pipe.  But my
> schedule (which seemed to magically fill up, ugh) makes it hard to do
> before next week.  In the meantime, check your traceroute to
> twinlark.arctic.org and see if it's satisfactory... hyperreal will be on
> the same pipes. 
> 
> Dean
> 
> On Thu, 26 Jun 1997, Rodent of Unusual Size wrote:
> 
> >     Judging by Paul's message, I'm not the only one having trouble with
> >     CVS.  So I infer that the problems are being caused by Hyperreal's
> >     router's knit-one-knot-one-drop-a-stitch, eenie-meenie-mynie packet
> >     discard setting.
> > 
> >     Brian, has your throughput improved (with the offload of the mirrors
> >     and mail) to the point that you can make a case for restoring full
> >     packet-pass functionality?
> > 
> >     #ken    :-P}
> > 




Re: Hyperrouter's Reality

Posted by Dean Gaudet <dg...@arctic.org>.
Wired's T3 is from BBN Planet.  They've only got 6 Mbits to MCI (and 3 to
sprint).  So the only major provider with poor coverage is uunet. 

Dean

On Fri, 27 Jun 1997, Sameer Parekh wrote:

> connects great over mci, fwiw. Can you get BBN to route to Wired
> through MCI, rather than UUNet?


Re: Hyperrouter's Reality

Posted by Sameer Parekh <sa...@c2.net>.
-----BEGIN PGP SIGNED MESSAGE-----

connects great over mci, fwiw. Can you get BBN to route to Wired
through MCI, rather than UUNet?

At 02:43 PM 6/26/97 -0500, Randy Terbush wrote:
>27% packet loss from here. The BBN-UUNET border router situation 
>does not look good from here.
>
>zab:/var/users/doc/www.sportdoc.com
>#> ping -s 1500  twinlark.arctic.org
>PING twinlark.arctic.org (204.62.130.91): 1500 data bytes
>1508 bytes from 204.62.130.91: icmp_seq=0 ttl=55 time=150.896 ms
>1508 bytes from 204.62.130.91: icmp_seq=1 ttl=55 time=150.455 ms
>1508 bytes from 204.62.130.91: icmp_seq=2 ttl=55 time=146.558 ms
>1508 bytes from 204.62.130.91: icmp_seq=3 ttl=55 time=136.520 ms
>1508 bytes from 204.62.130.91: icmp_seq=5 ttl=55 time=139.806 ms
>1508 bytes from 204.62.130.91: icmp_seq=8 ttl=55 time=148.004 ms
>1508 bytes from 204.62.130.91: icmp_seq=9 ttl=55 time=133.583 ms
>1508 bytes from 204.62.130.91: icmp_seq=10 ttl=55 time=137.575 ms
>1508 bytes from 204.62.130.91: icmp_seq=11 ttl=55 time=135.580 ms
>1508 bytes from 204.62.130.91: icmp_seq=12 ttl=55 time=146.005 ms
>1508 bytes from 204.62.130.91: icmp_seq=14 ttl=55 time=146.152 ms
>1508 bytes from 204.62.130.91: icmp_seq=16 ttl=55 time=144.001 ms
>1508 bytes from 204.62.130.91: icmp_seq=17 ttl=55 time=133.629 ms
>^?
>--- twinlark.arctic.org ping statistics ---
>18 packets transmitted, 13 packets received, 27% packet loss
>round-trip min/avg/max = 133.583/142.212/150.896 ms
>
>
>As opposed to my connection to taz:
>
>#> ping -s 1500  apache.org
>PING apache.org (204.152.144.38): 1500 data bytes
>1508 bytes from 204.152.144.38: icmp_seq=0 ttl=249 time=105.282 ms
>1508 bytes from 204.152.144.38: icmp_seq=1 ttl=249 time=106.956 ms
>1508 bytes from 204.152.144.38: icmp_seq=2 ttl=249 time=105.661 ms
>1508 bytes from 204.152.144.38: icmp_seq=3 ttl=249 time=111.128 ms
>1508 bytes from 204.152.144.38: icmp_seq=4 ttl=249 time=108.928 ms
>1508 bytes from 204.152.144.38: icmp_seq=5 ttl=249 time=108.576 ms
>1508 bytes from 204.152.144.38: icmp_seq=6 ttl=249 time=103.666 ms
>1508 bytes from 204.152.144.38: icmp_seq=7 ttl=249 time=105.187 ms
>1508 bytes from 204.152.144.38: icmp_seq=8 ttl=249 time=121.232 ms
>1508 bytes from 204.152.144.38: icmp_seq=9 ttl=249 time=120.758 ms
>1508 bytes from 204.152.144.38: icmp_seq=10 ttl=249 time=103.432 ms
>1508 bytes from 204.152.144.38: icmp_seq=11 ttl=249 time=112.013 ms
>1508 bytes from 204.152.144.38: icmp_seq=12 ttl=249 time=110.520 ms
>1508 bytes from 204.152.144.38: icmp_seq=13 ttl=249 time=103.341 ms
>1508 bytes from 204.152.144.38: icmp_seq=14 ttl=249 time=107.021 ms
>^?
>--- apache.org ping statistics ---
>15 packets transmitted, 15 packets received, 0% packet loss
>round-trip min/avg/max = 103.341/108.913/121.232 ms
>
>
>> Brian and I have arranged to move hyperreal to another pipe.  But 
my
>> schedule (which seemed to magically fill up, ugh) makes it hard to 
do
>> before next week.  In the meantime, check your traceroute to
>> twinlark.arctic.org and see if it's satisfactory... hyperreal will 
be on
>> the same pipes. 
>> 
>> Dean
>> 
>> On Thu, 26 Jun 1997, Rodent of Unusual Size wrote:
>> 
>> >     Judging by Paul's message, I'm not the only one having 
trouble with
>> >     CVS.  So I infer that the problems are being caused by 
Hyperreal's
>> >     router's knit-one-knot-one-drop-a-stitch, eenie-meenie-mynie 
packet
>> >     discard setting.
>> > 
>> >     Brian, has your throughput improved (with the offload of the 
mirrors
>> >     and mail) to the point that you can make a case for 
restoring full
>> >     packet-pass functionality?
>> > 
>> >     #ken    :-P}
>> > 
>
>
>
>
-----BEGIN PGP SIGNATURE-----
Version: PGP for Personal Privacy 5.0
Charset: noconv

iQEVAwUBM7RPA21/NmvdkZwxAQFH5QgAq6cvJ5SG2wTGzONWt5OHGUoI9u2dNO/u
WdmyCsJKXiOlRHyU1dv/8CAku5fSG5/tgPzJbAt/HHK9ngzuRJ//mgCPb79rINqW
reqmJ5IXmWdCbqzRT2t3Q6khTbpf0kUxyNp4ZCZHnh7ez54XX6YP6aZtueX7UQps
wFtXWPjfYRiBeH3Whj8ZeIXX1GBmjwefXU3+oOJab+sBE1e25o56fIHn5l+gWQo0
SrmzGXJZLAx06uKsVBKMvK/BHj8gsBB7xZegJ73OP0zSZ4s6K+oKDijFAsGG4Hfq
alc2WNrfci8Z/y8xiwesejz08mfApLXaEiyNgfzcbPzDAf3QfnP47g==
=soJ7
-----END PGP SIGNATURE-----